rlaope/oh-my-hermes

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Local scripts → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Local scripts

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Local scripts → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Local scripts

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Local scripts → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Local scripts

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Network or broad shell → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Network or broad shell

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Local scripts → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Local scripts

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Network or broad shell → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Network or broad shell

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Local scripts → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Local scripts

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → 98284f305f6a

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Network or broad shell → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Network or broad shell

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 60 - If the user moves from terminology to implementation, summarize confirmed understanding and hand off to `ralplan`, `ulw-plan`, or the selected coding owner only after a separate go-ahead.

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 65 - If current-source evidence is missing, route a `research` step before accepting the plan.

This preview is limited. See the complete change on GitHub.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

- 49 - If the expected behavior is unclear, route back to plan before running adversarial checks.
- 50 - If verification fails, return to fix or research with the failed signal instead of advancing.
+ 49 - If a scenario needs a real install or external service the harness cannot reach, retain other observed pass/fail evidence, mark that scenario not_observed, and hand back its manual test guide with setup, do, expect, and broken signals.
+ 50 - If a hostile browser scenario would submit, buy, send, or delete on a live target without approval, retain read-only scenario results, mark that action not_observed, and hand back the target and exact action for user approval.
+ 51 - If a failing browser scenario has no screenshot or console readback, keep its observed failure, mark diagnosis tentative, and hand back the scenario, target, and missing capture for a bounded rerun.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 93 - When Hermes' bundled `grounded-citations` skill is available, load it and register each source in its ledger when it is retrieved (`scripts/sources.py add`). Keep the run's ledger in a file of its own, passed as `--ledger <path>` before the subcommand on every call: the default ledger is shared by the whole profile, and another run's reset rebinds its ids. Use parent-only serialized writes: lanes return observed source entries and quote refs; the parent registers URLs, attaches quotes, and remaps lane ids before drafting. Lanes never write the shared ledger. When a human briefing is written, report the citation check from a run, or as `not_observed`: the exit status of `sources.py verify <draft>` is the observed check - `0` is a pass and any other status is not - and `verify --evidence` also fails a draft whose cited source carries no verbatim quote. An exit `0` whose `info: stats:` line reads `0 distinct source(s) cited` checked nothing and is reported as `not_observed`. It is a citation check, not fact verification, and it never stands in for the contested-claim gate. Without the skill, keep each source's URL and retrieval date in the research notes and report the citation check as `not_observed`.
+ 93 - When Hermes' bundled `grounded-citations` skill is available, load it and register each source in its ledger when it is retrieved (`scripts/sources.py add`). Keep the run's ledger in a file of its own, passed as `--ledger <path>` before the subcommand on every call: the default ledger is shared by the whole profile, and another run's reset rebinds its ids. Use parent-only serialized writes: lanes return observed source entries and quote refs; the parent registers URLs, attaches quotes, and remaps lane ids before drafting. Lanes never write the shared ledger. When a human briefing is written, report the citation check from a run, or as `not_observed`: the exit status of `sources.py verify <draft>` is the observed check - `0` is a pass and any other status is not - and `verify --evidence` also fails a draft whose cited source carries no verbatim quote. An exit `0` whose `info: stats:` line reads `0 distinct source(s) cited` checked nothing and is reported as `not_observed`. It is a citation check, not fact verification, and it never stands in for the contested-claim gate. After that verify, run `omh research receipts verify --citations <ledger> --draft <draft>` and report its fetched coverage from the run, or as `not_observed`; a receipt proves only that the page was opened in this session tree, never that it supports the claim. Without the skill, keep each source's URL and retrieval date in the research notes and report the citation check as `not_observed`.
- 96 - Decompose the question into orthogonal research axes and disambiguate named entities before any deep reading.
+ 96 - Decompose the question into orthogonal research axes and disambiguate named entities before any deep reading. Before answering from a source, confirm it studies the exact entity the question names; a different name is a different entity unless a source says they are the same, so a source on a related one is a lead, not an answer.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 27 - Plan the graph and call `omh_dag action=define` with the complete run `goal` and each node's id, label, category, prompt, depends_on and write_scope; fix every refusal and read the lint before going on. Never `delegate_task` before `define` returns its calls. The saved goal reaches every native task; local VERIFY does not replace goal acceptance.
+ 27 - When the conversation holds a plan the user accepted (from `plan`, `ralplan` or `refactor-plan`), build the graph from it instead of re-planning: each task becomes a node with its id, depends_on and write scope, its done check goes into VERIFY and STOP WHEN, and its closing verification task becomes the `role: verify` node; plan only what it leaves open.
+ 28 - Plan the graph and call `omh_dag action=define` with the complete run `goal` and each node's id, label, category, prompt, depends_on, write_scope and role (`build` by default; `baseline`, `verify`, `review`); fix every refusal and read the lint before going on. Never `delegate_task` before `define` returns its calls. The saved goal reaches every native task; local VERIFY does not replace goal acceptance.
- 91 - Climb the difficulty ladder `quick` -> `unspecified-low` -> `unspecified-high` with a one-line reason per climb; use `ultrabrain` at most once, and a graph whose every node is `deep` or `deep-work` is a routing failure.
+ 92 - Climb the difficulty ladder `quick` -> `unspecified-low` -> `unspecified-high` with a one-line reason per climb; use `ultrabrain` at most once, and a graph whose every node is `deep` or `deep-work` is a routing failure. A `role=review` node takes category `review` where your routes serve it; a `role=baseline` node stays `quick` or `unspecified-low`; a `role=verify` node judges the run goal and baseline exemptions, so climb the ladder for it as for any judgement.
- 95 - End every run with a verification node that depends on every producer, runs the real command, and is never the node that built the work.
- 96 - Completion is a claim: before reporting the goal met, check each node against its SCOPE in both directions, nothing owed missing and nothing beyond it done.
- 97 - Goal loop: verify, repair, extend, verify. A failed check gets repair nodes (`omh_dag action=repair`, two cycles per node); when every node passed but the goal check finds work missing, add it to the same run (`omh_dag action=extend`, two rounds per run). Past either limit, report what is still open.
+ 96 - End every run with a `role: verify` node that depends on every producer, runs the real command, and is never the node that built the work.
+ 97 - Completion is a claim: report the goal met only when `omh_dag` says `verdict: verified`, and first check each node against its SCOPE in both directions, nothing owed missing and nothing beyond it done.
+ 98 - Goal loop: verify, repair, extend, verify. A failed check gets repair nodes (`omh_dag action=repair`, two cycles per node, the last re-plans); when every node passed but the goal check finds work missing, add it to the same run (`omh_dag action=extend`, two rounds per run). Past either limit, report what is still open.
+ 99 - Unknown path: scout first; see 'When the path is unknown' in `references/run-loop.md`.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

- 51 - If provider metrics are unavailable, report only local metadata and mark provider truth not_observed.
- 52 - If cost or latency looks risky, surface a warning plus the next measurement rather than a completion claim.
+ 51 - If neither active-profile nor legacy hermes-achievements files are available, no badge result stands; mark the summary not_observed and hand back the missing local-file evidence needed for hermes_achievements_observation/v1.
+ 52 - If the snapshot is older than the plugin's rescan window, retain its badge list as of generated_at, mark current progress tentative, and hand back the timestamp and stale-snapshot warning.
+ 53 - If supplied session_activity_receipt/v1 lacks a metric, retain the available badge and activity metadata, mark that metric not_observed, and hand back the gap without replacing it with zero.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 57 - If the proposal under review cannot be stated in one paragraph, route back to `deep-interview` before opening round one.
+ 57 - If the proposal under review cannot be stated in one paragraph, route back to `ulw-interview` before opening round one.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 107 - agent_debug_report/v1 from `omh quality-evidence agent-debug --hermes-session <id|latest> --observable <kind> --json` (or `--session-record <jsonl>`, `--turns N:M`): tool errors, identical retries after an error, background processes started without notify_on_complete, and compaction boundaries, each cited by session and message id or record line; treat a kind listed unavailable as unchecked
+ 107 - agent_debug_report/v1 from `omh quality-evidence agent-debug --hermes-session <id|latest> --observable <kind> --json` (or `--session-record <jsonl>`, `--turns N:M`): tool errors, identical retries after an error, background processes started without notify_on_complete, compaction boundaries, and silent turn ends (an empty assistant row right after a tool result that ended the turn), each cited by session and message id or record line; treat a kind listed unavailable as unchecked, and when a silent turn end leads, read the host's agent.log `Turn ended: reason=` line for its timestamp before changing anything

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 53 - If required context is missing, ask one blocking question or route back to the narrower workflow.
- 54 - If runtime or wrapper evidence is unavailable, keep the status as not_observed and expose the next observable action.
+ 53 - If observed_at lacks receipt-authenticated provenance, retain the task set, rubric, and input digests, mark the comparison not_observed, and hand back paired_run_decision/v1 with the provenance gap rather than a pass or fail.
+ 54 - If candidate inputs or success criteria differ outside the variable under test, retain the observed runs, mark executor selection tentative, and hand back the mismatched digests or criteria and a matched rerun budget.
+ 55 - If a judge has no measured agreement against human labels, retain the prewritten rubric and observed executable checks, mark judgment unresolved and unqualified, and hand back the missing agreement measurements instead of a correctness claim.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 51 - If a managed path or config key is missing, route to setup/update repair instead of editing hidden state.
- 52 - If a reload or plugin load was not observed, keep the diagnostic result as local health evidence only.
+ 51 - If only a prepared handoff exists, mark coding progress not_observed; hand back the missing stage.
+ 52 - If P0 capture fails its audit, keep the tier checks, mark readiness unresolved; hand back quick wins.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Local scripts → After Network or broad shell

d8697e46651e → 98284f305f6a

- 69 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 69 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Local scripts

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 55 - If the selected executor is unavailable, ask for Codex, Claude Code, Hermes, or another runtime before retrying.
- 56 - If dispatch or result evidence is missing, keep the handoff prepared_not_observed and expose the next observable action.
+ 55 - If baseline regression checks cannot run, retain the categorized smell inventory, mark the behavior lock unresolved, and hand back the cleanup plan and missing before-checks without starting risky edits.
+ 56 - If verification fails between single-smell passes, retain the last verified pass's evidence, mark the new pass unresolved, and hand back its small diff and failed check for a fix or revert before the next pass.
+ 57 - If the cleanup has no runtime record, retain its observed diff and before/after checks, mark the cleaner run not_observed, and hand back the four-part closing report for recording before an AC-SLOP checkpoint.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 57 - The fix handoff carries the reproduction as a regression test that fails before and passes after.
+ 57 - The fix handoff carries the reproduction as a regression test that fails before and passes after, or records `regression_seam: none_correct` with the reason, with the reproduction loop observed red before the fix and green after it.
+ 58 - An observed search for the session's probe tag returns nothing, and every throwaway harness file is deleted or listed.
- 61 - If the fault does not reproduce, make reproduction the first hypothesis and plan the loop, seed, or ordering that would establish it.
+ 62 - If the fault does not reproduce, make reproduction the first hypothesis and plan the loop, seed, or ordering that would establish it; when no test reaches the fault, choose a loop from the method reference's loop table; if none can be built, the record stays not_observed with each attempt and what stopped it, and the request to the user is for an environment, a redacted capture, or permission for tagged temporary instrumentation.
+ 89 - A reproduction counts for this bug only when its `symptom_match` shows the failure the user reported; a different failure near the same code is another bug, and this record stays not_observed.
+ 90 - `minimised` is filled before `competing_hypotheses/v1`: an input, caller, config value, or step leaves the reproduction only when the loop rerun without it still shows the symptom (for an intermittent fault, the same rate at the same N), and the record lists what left and what stayed; `fix_handoff/v1` builds the regression test from what stayed.
+ 95 - Before the fix handoff names a regression test, load `skill_view(name='omh-verification-gate', file_path='references/test-authoring.md')` for the seam, the shapes of tests that guard nothing, fakes only at system boundaries, and the no-correct-seam outcome.
- 117 - reproduction_record/v1 names the exact command, the input, the observed output, the expected output, and the hit rate over N runs; it reads observed or not_observed and nothing between
+ 121 - reproduction_record/v1 names the exact command, the input, the observed output, the expected output, and the hit rate over N runs, plus `symptom_match` (the observed failure quoted beside the reported one), `loop_cost` (wall time per run, and what was pinned or, for an intermittent fault, how the rate was raised), and `minimised` (the cuts made and the load-bearing form left); it reads observed or not_observed and nothing between
- 120 - root_cause_record/v1 cites the observation that demonstrated the cause and the one that ruled out each rival
- 121 - fix_handoff/v1 carries the reproduction as the regression test that must fail before the fix and pass after it
+ 124 - root_cause_record/v1 cites the observation that demonstrated the cause and the one that ruled out each rival, and the fix's commit or pull-request body states that confirmed hypothesis and the rivals it eliminated, because the next debugger reads history, not this record
+ 125 - fix_handoff/v1 carries the reproduction as the regression test that must fail before the fix and pass after it, or records `regression_seam: none_correct` with the reason, with the reproduction loop observed red before the fix and green after it
+ 126 - fix_handoff/v1 names the seam the regression test drives and how it runs the pattern that produced the bug (callers, ordering, call chain); when the pattern cannot run at any reachable seam, it records `regression_seam: none_correct` with the reason and what was tried, keeps the reproduction loop as the fix's verification, and reports the missing seam as an `omh-refactor-plan` candidate instead of a shallow test that passes before and after the fix
+ 135 - Redact before recording: reproduction commands read credentials from environment variables by name, quoted output keeps only the lines that carry the signal with each secret replaced by `<redacted>`, and a capture too redacted to diagnose is reported as such, never recorded unredacted.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

- 59 - If the request turns out to be about the agent's own prompts, tools, or credentials, stop and hand it to `security-safety-review`.
+ 59 - If the request turns out to be about the agent's own prompts, tools, or credentials, stop and hand it to `omh-security-safety-review`.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 52 - If the reviewed target is missing, inspect the requested artifact or ask one target question.
- 53 - If independent verification is unavailable, report the gap and avoid an approval-style claim.
+ 52 - If no advisor is configured, retain the question and context summary, mark external advice not_observed, and hand back the agent's own answer without choosing a default advisor.
+ 53 - If the named advisor returns no response, retain the prepared advisor prompt, mark external advice not_observed, and hand back the question and context summary for a later consultation.
+ 54 - If advisor claims lack file, artifact, command, or source evidence, retain the captured advice as advice only, mark those conclusions tentative, and hand back accepted/rejected points and unresolved questions in the decision note.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 55 - If required context is missing, ask one blocking question or route back to the narrower workflow.
- 56 - If runtime or wrapper evidence is unavailable, keep the status as not_observed and expose the next observable action.
+ 55 - If a stop condition is missing, retain the recurring request and cadence, mark the choice among cron, heartbeat, loop, and native goal unresolved, and hand back recurring_surface_comparison/v1 with the stop-condition question.
+ 56 - If overlap, missed-run, retry, backfill, or failure-pause decisions remain open, retain prepared_native_blueprint/v1, mark activation unresolved, and hand back the missing policy choices and native-runtime limits before activation.
+ 57 - If no run reference exists for the saved intent revision, retain hermes_recurring_intent/v1, mark execution and delivery not_observed, and hand back `hermes cron runs` and `hermes cron incidents` as the next status checks.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

+ 90 - Name the guard on every write two requests can reach at once — a unique constraint, a stored idempotency key, a lost-update guard — and send side effects only after commit; the pagination-cursor, integrity, timeout, and cache rules are in sections 4 to 6 of the service-contract reference.
+ 91 - When the surface receives webhooks, consumes a queue, or sizes a connection pool, load `references/inbound-and-async.md`: verify a webhook over its raw body before parsing, and treat every queued message as one that can arrive more than once.
+ 93 - When the handoff asks for tests, load `references/contract-tests.md` and derive the test plan in backend_implementation_handoff/v1 from the contract: one negative test per auth_boundary_map/v1 row, one per error_path_table/v1 row, stand-ins chosen by dependency category with the same database engine, and migrations run on populated data.
- 120 - backend_service_contract/v1 names each endpoint or job, its caller class, request and response shapes, and its idempotency and pagination rules
+ 123 - backend_service_contract/v1 names each endpoint or job, its caller class, request and response shapes, its idempotency and pagination rules, and its expected rate, latency target, and payload bound (or `unknown`)
+ 136 - Treat an inbound webhook handler that does not verify the signature over the raw body before parsing as a blocker, not a follow-up.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Local scripts → After Network or broad shell

d8697e46651e → 98284f305f6a

- 70 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 70 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Local scripts

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 63 - For maintainability judgement calls, load `omh-code-review/references/smell-baseline.md`; it names the twelve baseline smells with their fixes and the repo-standards-override rule.
+ 63 - For maintainability judgement calls, load `omh-code-review/references/smell-baseline.md`; it finds the repository's standards files first, then names the twelve baseline smells with their fixes and the repo-standards-override rule.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 46 - The plan names goals, non-goals, assumptions, acceptance criteria, and verification shape.
- 47 - Draft recommendations, accepted decisions, and executor handoffs are separate states.
- 48 - Rejected options or unresolved tradeoffs are recorded before handoff.
+ 46 - repo_map/v1 cites an observed path for every directory, entrypoint, generated surface, and test location it names; what was inferred is marked inferred, and what was never opened is listed as unread.
+ 47 - reading_path/v1 is an ordered list of real files, each with why it comes at that point and what the reader knows after it.
+ 48 - Every domain_glossary/v1 term points to where it is defined or used; a term with no source is an open question, not an entry.
+ 49 - risk_and_unknowns_map/v1 keeps observed risks, each with the file or command that shows it, apart from unknowns, each with how it would be resolved.
+ 50 - Every first_task_runway/v1 task names the files it touches and its verification command, and says whether that command was run during onboarding.
+ 51 - The stop condition named at the start is met, or the reply names what was left unread.
- 52 - If acceptance criteria or verification are missing, route back to clarification before handoff.
- 53 - If assumptions materially affect the plan, keep them visible and avoid treating the plan as accepted.
+ 55 - If the repository is too large for the stated depth, narrow to the entrypoint the audience touches first and record the rest as unread rather than summarizing it from file names.
+ 56 - If a build, test, or setup command could not run (read-only mode, missing toolchain, secrets), list it in first_task_runway/v1 as not run with the reason, never as a known-good command. A command that installs or syncs dependencies as a side effect (`npm ci`, `uv sync`, and `uv run`, which syncs first) counts as setup and stays not run.
+ 57 - If docs and code disagree, report what the code does, cite both, and add the disagreement to risk_and_unknowns_map/v1 instead of trusting the docs.
+ 58 - If the user names a concrete task mid-tour, stop at the current reading_path/v1 entry and hand the runway so far to `omh-plan`, or to `ulw-work` or `omh-idea-to-deploy` when the task already has acceptance criteria.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 52 - If provider metrics are unavailable, report only local metadata and mark provider truth not_observed.
- 53 - If cost or latency looks risky, surface a warning plus the next measurement rather than a completion claim.
+ 52 - If the previous must-keep pack has no item classes, retain its recorded contents, mark the class delta unresolved, and hand back must_keep_item_class_delta/v1 as comparison unavailable, never as zero items.
+ 53 - If a checkpoint is due but no Hermes /compress reply is observed, retain the must-keep context pack and checkpoint plan, mark compaction not_observed, and hand back /compress <focus> with the reply needed to verify continuity.
+ 54 - If the session has no published context_budget_plan/v1, retain the proposed budget plan, mark inherited capacity unresolved, and hand back the capacity_unknown_hold and session reference for publication or rebinding.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 50 - If required context is missing, ask one blocking question or route back to the narrower workflow.
- 51 - If runtime or wrapper evidence is unavailable, keep the status as not_observed and expose the next observable action.
+ 50 - If operating signals do not support an architecture tradeoff, retain the cited signals, mark that tradeoff tentative, and hand back the decision note with assumptions and delivery risks.
+ 51 - If the decision owner has not accepted implementation follow-ups, retain the priority frame and risk ranking, mark follow-up authority unresolved, and hand back the decision note and handoff candidates for acceptance.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 53 - If acceptance criteria or verification are missing, route back to clarification before handoff.
- 54 - If assumptions materially affect the plan, keep them visible and avoid treating the plan as accepted.
+ 53 - If an outcome lacks acceptable assessment evidence, retain curriculum_learner_outcome_brief/v1, mark curriculum_alignment_map/v1 tentative, and hand back the orphan outcomes or assessment mismatches for revision.
+ 54 - If an access or technology barrier has no equivalent demonstration path, retain the aligned outcomes and assessments, mark curriculum_sequence_design/v1 unresolved, and hand back the barrier and required adaptation.
+ 55 - If learner review or pilot evidence is absent, return the alignment and sequence as unvalidated drafts, mark validation unresolved, and hand back curriculum_validation_disposition/v1 with required reviews and revalidation checks.
- 58 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 59 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 58 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 58 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 67 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 67 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 52 - If evidence is mostly assumption, label it and recommend a research or feedback-triage pass.
- 53 - If the decision owner is missing, keep the output as options rather than accepted strategy.
+ 52 - If demand and capacity cannot be compared in one unit or hiring ramp lag is unknown, retain the stated constraints and options, mark the recommendation tentative, and hand back the missing capacity and timing inputs.
+ 53 - If a drafted decision record lacks write approval, retain the options, tradeoffs, and proposed decision note, mark persistence not_observed, and hand back the draft for approval without writing docs/adr/.
- 57 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 57 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

- 18 Required: one decision, a bounded experiment, an isolated scratch boundary, and a measurement method. Output: a decision record, prepared handoff, observation ledger, and receipt. When Hermes' bundled `spike` skill is loaded for the same question, build the way it says and keep these records: `references/spike-composition.md`.
+ 18 Required: one decision, a bounded experiment, an isolated scratch boundary, and a measurement method. Output: a decision record, prepared handoff, observation ledger, and receipt. When Hermes' bundled `spike` skill is loaded for the same question, build the way it says and keep these records: `references/spike-composition.md`. For a logic, state, or layout question, or prototype code a later plan may reference, load `references/artifact-shapes.md`.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 51 - If required context is missing, ask one blocking question or route back to the narrower workflow.
- 52 - If runtime or wrapper evidence is unavailable, keep the status as not_observed and expose the next observable action.
+ 51 - If the query, scope, tags, stale policy, or match limit is unspecified, no rejected-decision match stands yet; mark recall unresolved and hand back one question fixing the missing filter.
+ 52 - If a candidate's reviewed-rejected status or expiry cannot be checked, retain only matches that passed those checks, mark remaining recall unresolved, and hand back rejected_decision_recall/v1 with the claim boundary and missing checks.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 51 - If generation tooling is missing, prepare a prompt or package handoff and mark file output not_observed.
- 52 - If QA or attachment evidence is missing, keep generated/delivered states separate and show the next check.
+ 51 - If rendered images do not match slide_count or remain uninspected, retain the observed slide_count, mark render_screenshot unresolved, and hand back the exact file and outstanding image checks.
+ 52 - If the person's approval names no final file, retain each generated version's read-back, mark approval unresolved, and hand back the stable id and version choices without selecting by modification time.
+ 53 - If a publication URL resolves to a different content hash, retain the approved file and any archive evidence already read back, mark publication unresolved, and hand back the URL and expected sha256 for destination read-back.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 52 - If required context is missing, ask one blocking question or route back to the narrower workflow.
- 53 - If runtime or wrapper evidence is unavailable, keep the status as not_observed and expose the next observable action.
+ 52 - If a monitor has only its first baseline tick or a silent tick, retain the checklist and watchlist, mark release health not_observed, and hand back the recorded cron runs and open incidents needed to assess health.
+ 53 - If the rollback owner is missing, retain the release scope and health signals, mark the deploy decision gate unresolved, and hand back the rollback criteria with the owner question before advancing.
+ 54 - If a canary lacks host_deployment_observation/v1 or a pre-deploy production baseline, retain the checklist, mark the comparison unresolved and BLOCK, and hand back the missing observation or baseline without authorizing rollback.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 53 - If the selected executor is unavailable, ask for Codex, Claude Code, Hermes, or another runtime before retrying.
- 54 - If dispatch or result evidence is missing, keep the handoff prepared_not_observed and expose the next observable action.
+ 53 - If the scratch-edit probe is denied, keep the denial, mark permission unresolved; hand back the grant needed.
+ 54 - If a dispatch result has no session_id or thread_id, mark resumability unresolved; hand back the JSON flag.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 54 - If feedback lacks source or severity, ask for the missing signal before coding handoff.
- 55 - If the item is actually a plan or research request, route to that workflow instead of triage.
+ 54 - If no feedback items are supplied, no cluster stands; mark triage not_observed and hand back the request for feedback items or a summary, source boundary, and product area.
+ 55 - If bug and feature signals cannot be separated from the supplied reports, retain source-bounded clusters, mark their classification and ranking tentative, and hand back the missing evidence with research or strategy follow-up options.
+ 56 - If a triaged cluster has no explicit coding intent or accepted-plan evidence, retain the clusters and ranking, mark the coding handoff unresolved, and hand back the next workflow recommendation with product_evidence_loop/v1 still prepared-only.
- 59 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 60 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 55 - If required context is missing, ask one blocking question or route back to the narrower workflow.
- 56 - If runtime or wrapper evidence is unavailable, keep the status as not_observed and expose the next observable action.
+ 55 - If period, currency, basis, or comparator version conflicts, retain the supplied source provenance, mark comparability unresolved, and hand back finance_scope_source_record/v1 with the conflicting attributes.
+ 56 - If totals or account mappings do not tie out, retain the confirmed scope and source record, mark reconciliation unresolved, and hand back finance_reconciliation_analysis_schedule/v1 with unreconciled gaps.
+ 57 - If recalculation fails at brief validation, retain supported calculations and finance_risk_register/v1, mark finance_decision_brief/v1 unresolved with HOLD, and hand back the failed calculation and escalation owner.
- 60 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 61 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 51 - The frontend_route_state_matrix/v1 covers pages, 375/768/1280-style breakpoints, empty/loading/error, interaction, and CJK/locale risks.
+ 51 - The frontend_route_state_matrix/v1 covers pages, 375/768/1280-style breakpoints, empty/loading/error, interaction, and CJK/locale risks, plus submitting, submit-failed, optimistic-pending, stale/refetching, and conflict rows for remote writes.
+ 95 - When the surface reads or writes remote data, load the frontend-refactor skill's `omh-frontend-refactor/references/state-discipline.md` (sections 1 and 4-6) before writing the component: server state lives in a query cache, each mutation names the query keys it invalidates, and an optimistic write ships with its rollback.
+ 96 - When the surface has a form, a text composer, or a dialog, load `references/forms-and-input.md` before writing it: labelled controls, errors shown after the user commits a value, server field errors mapped onto their fields, one submit in flight, an Enter-to-send that waits for IME composition, and focus returned when a dialog closes.
+ 97 - Check the browser-support target before using a newer platform feature: a feature newer than the target (by default, anything only Baseline Newly available) needs feature detection plus a fallback, or a deliberate target change; suggest recording the target in the repository's agent instruction file (`omh-agent-instructions` owns that file).
+ 104 - When the handoff asks for component or interaction tests, load `skill_view(name='omh-verification-gate', file_path='references/test-authoring.md')` and name the seam each test drives: assert through what a user perceives (role, accessible name, visible text) across the frontend_component_state_inventory/v1 states, never class names or component-internal state, and fake only at the boundaries the shared page lists (network, clock, randomness).
+ 121 - browser-support target (default: Baseline Widely available, in every core browser for 30+ months)
- 140 - frontend_route_state_matrix/v1 with pages, states, viewports, CJK/locale, empty/loading/error, and interaction states
- 141 - frontend_component_state_inventory/v1 with default, hover, focus, active, disabled, loading, empty, and error states for reusable primitives
+ 145 - frontend_route_state_matrix/v1 with pages, states, viewports, CJK/locale, empty/loading/error, and interaction states, plus submitting, submit-failed, optimistic-pending, stale/refetching, and conflict (409) rows wherever the surface writes remote data
+ 146 - frontend_component_state_inventory/v1 with default, hover, focus, active, disabled, loading, empty, and error states for reusable primitives, and each primitive's interface: props, events, slots, and the invariants a caller must know; a wrapper earns a row only when deleting it would push its complexity into its callers

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 84 - Gate macro changes on characterization tests written before the refactor; snapshot tests lock markup, not behavior, and do not count.
+ 84 - Gate macro changes on characterization tests written before the refactor; snapshot tests lock markup, not behavior, and do not count. A change that alters DOM nesting, element type, or class composition also needs a before/after capture at one viewport, because green tests do not prove the render is unchanged.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 51 - If platform metadata is missing, keep the card platform-neutral and ask for the target surface.
- 52 - If send or registration evidence is unavailable, show the adapter-owned action instead of claiming delivery.
+ 51 - If silent update conflicts with a status notice, keep the thread; hand back the silent-or-notify choice.
+ 52 - If an action would touch a thread other than the named one, mark it unresolved; hand back the target.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Local scripts → After Network or broad shell

d8697e46651e → 98284f305f6a

- 69 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 69 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Local scripts

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 54 - If required context is missing, ask one blocking question or route back to the narrower workflow.
- 55 - If runtime or wrapper evidence is unavailable, keep the status as not_observed and expose the next observable action.
+ 54 - If reproduction evidence is still missing after three questions, retain the report classification and observed facts, mark the direction unresolved, and hand back the missing steps, versions, or logs without filing.
+ 55 - If duplicate search cannot finish, retain the scoped report summary, mark duplicate status unresolved, and hand back the explicit repository and pending search; do not confirm or dispatch creation.
+ 56 - If connector read-back lacks request binding or issue fields, retain the confirmed direction and duplicate status, mark creation not_observed, and hand back the idempotency key and missing read-back fields without redispatching.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 51 - If required context is missing, ask one blocking question or route back to the narrower workflow.
- 52 - If runtime or wrapper evidence is unavailable, keep the status as not_observed and expose the next observable action.
+ 51 - If the success metric or release scope is missing, retain the supplied idea and user value, mark delivery planning unresolved, and hand back the stage rail with the missing decision before an executor handoff.
+ 52 - If plan acceptance or executor choice is absent, retain the prepared decision gates and stage rail, mark the coding handoff unresolved, and hand back the acceptance or runtime-choice question.
+ 53 - If deploy evidence is missing after verification, retain the observed verification results, mark deploy and monitoring not_observed, and hand back the stage rail with the deploy evidence gate still open.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 52 - If a renderer or file tool is missing, keep the package prepared and expose the generation handoff.
- 53 - If render QA is unavailable, mark the artifact unverified and request the smallest visual/file check.
+ 52 - If image_generation_capability/v1 is unknown or prompt_only, retain visual_prompt_card/v1, mark generation not_observed, and hand back image_generation_setup/v1 with the question of which image tool to use.
+ 53 - If an edit lacks the supplied image reference or preserve/remove constraints, retain only the supplied edit intent, mark transformation unresolved, and hand back the missing inputs for the image transformation brief.
+ 54 - If visual_generation_receipt/v1 reports a failed or partial attempt, retain visual_prompt_card/v1 and the receipt's failure stage, mark generation not_observed, and hand back the failed attempt without claiming an image.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Network or broad shell → After Network or broad shell

d8697e46651e → 98284f305f6a

- 60 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 60 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Network or broad shell

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 65 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 65 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Network or broad shell → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Network or broad shell

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 64 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 64 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 54 - If the reviewed target is missing, inspect the requested artifact or ask one target question.
- 55 - If independent verification is unavailable, report the gap and avoid an approval-style claim.
+ 54 - If an authority's pinpoint or effective status cannot be verified, retain the established document scope, mark authority-dependent issues unresolved, and hand back legal_scope_authority_record/v1 with the missing citations for counsel.
+ 55 - If a contested clause has an open counsel hold or no cited playbook position, retain its issue row, mark that redline row unresolved, and hand back the counsel question rather than proposed language.
+ 56 - If mandatory counsel triggers remain open at final review, retain the issue matrix and hold register, mark legal_review_disposition/v1 unresolved with HOLD, and hand back the exact triggers and counsel owner without a final determination.
- 59 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 60 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

- 59 - If the incident turns out to be closed, hand the postmortem to `reliability-review` and leave this record as the timeline it reads.
+ 59 - If the incident turns out to be closed, hand the postmortem to `omh-reliability-review` and leave this record as the timeline it reads.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

+ 53 - The client boundary has a deterministic fake adapter, the code around the model is unit-tested against it without a network call, and a streamed call is cancelled upstream when its client goes away.
- 91 - Route every provider call through one client boundary module that owns the model ID, credentials, timeout, retry policy, and rate-limit backoff. A second call site that builds its own client is how a model pin, a timeout, and a retry policy quietly diverge.
+ 92 - Route every provider call through one client boundary module that owns the model ID, credentials, timeout, retry policy, and rate-limit backoff. A second call site that builds its own client is how a model pin, a timeout, and a retry policy quietly diverge. Give that boundary a deterministic fake adapter beside the real one, so the code around the model is unit-tested offline; evals against the real model are a separate suite.
- 93 - Take structured output from a declared schema - a JSON schema, a typed parser, or the provider's structured-output mode - and validate every response against it. A response that fails validation is repaired by one bounded re-ask that shows the validation error, then fails loudly; it is never regex-scraped out of prose.
+ 94 - Take structured output from a declared schema - a JSON schema, a typed parser, or the provider's structured-output mode - and validate every response against it. Read the stop reason first: a response cut off at the output cap is truncated, not invalid, and is never re-asked under the same cap. A response that fails validation is repaired by one bounded re-ask that shows the validation error, then fails loudly; it is never regex-scraped out of prose.
- 98 - Give every agentic loop its budgets as product features, not prompt advice: step, time, token, cost, and tool-call budgets each with a recorded termination reason, and for recursive delegation the budgets bind the whole tree, not each node separately.
+ 99 - Give every agentic loop its budgets as product features, not prompt advice: step, time, token, cost, and tool-call budgets each with a recorded termination reason, and for recursive delegation the budgets bind the whole tree, not each node separately. Tool arguments are validated before the tool runs, and an invalid call or a tool failure returns to the model as a tool result, never thrown past the loop and never swallowed.
- 132 - Do not catch provider failures broadly; classify timeout, rate limit, transient server error, invalid request, and content refusal separately, because only some of them are safe to retry.
+ 133 - Do not catch provider failures broadly; classify timeout, rate limit, transient server error, invalid request, content refusal, and truncation at the output cap separately, because only some of them are safe to retry.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 53 - If the reviewed target is missing, inspect the requested artifact or ask one target question.
- 54 - If independent verification is unavailable, report the gap and avoid an approval-style claim.
+ 53 - If the source version cannot be established, retain supplied strings and the locale/audience brief, mark terminology and cultural-fit choices tentative, and hand back the source-version question before treating the glossary as approved.
+ 54 - If rendered UI evidence is missing, retain glossary choices and the string issue matrix, mark rendered validation not_observed, and hand back locale QA criteria and review owners for an accepted `omh-visual-qa` handoff.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 64 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 64 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 68 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 68 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 53 - If a renderer or file tool is missing, keep the package prepared and expose the generation handoff.
- 54 - If render QA is unavailable, mark the artifact unverified and request the smallest visual/file check.
+ 53 - If source metrics needed by an outline section are missing, retain the verified source facts and outline, mark that section unresolved, and hand back material_artifact/v1 with its missing inputs and output owner.
+ 54 - If xlsx_recalc.py reports recalculated false, retain the observed sheet_schema, mark formula_recalculation not_observed, and hand back its missing tools and the --out <copy> recalculation and cached-value read-back steps.
+ 55 - If DOCX package validation passes without a rendered converted PDF, retain observed section_structure and required_clause_check, mark appearance and page_count not_observed, and hand back the converted-PDF preview checks.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 53 - If participants, purpose, or decision owner are missing, ask for the one field that changes the agenda.
- 54 - If minutes or decisions were not observed, keep the output as prep rather than record.
+ 53 - If decision topics lack known context, retain the supplied meeting goal and audience, mark the agenda tentative, and hand back discussion prompts and the decisions-needed list with its open questions.
+ 54 - If action owners, due dates, or acceptance conditions are absent from a meeting record, retain the prepared agenda and prompts, mark action ownership not_observed, and hand back the action-item template with those columns marked unknown.
+ 55 - If follow-up coding lacks observed decisions or an accepted plan, retain the meeting brief, mark the coding handoff unresolved, and hand back the decisions needed rather than an implementation claim.
- 58 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 59 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 51 - If required context is missing, ask one blocking question or route back to the narrower workflow.
- 52 - If runtime or wrapper evidence is unavailable, keep the status as not_observed and expose the next observable action.
+ 51 - If project versus user scope or lasting versus temporary retention is ambiguous, retain the bounded summary as a candidate, mark capture unresolved, and hand back one scope-or-durability question without claiming a write.
+ 52 - If omh_memory returns error, retain memory_new_candidate/v1 as prepared only, mark the write not_observed, and hand back the bounded summary, record_type, tags, scope, and retention_class for the omh memory capture fallback.
+ 53 - If capture returns pending_review, retain the staged memory_new_candidate/v1, mark approval not_observed, and hand back the stated review reason without treating an unsafe or relative-time phrase as an approved record.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 52 - If required context is missing, ask one blocking question or route back to the narrower workflow.
- 53 - If runtime or wrapper evidence is unavailable, keep the status as not_observed and expose the next observable action.
+ 52 - If entry-by-entry confirmation stops before the assembled diff is approved, retain the reviewed claim decisions, mark the diff unresolved, and hand back reviewed indexes, remaining candidates, and the next entry to resume.
+ 53 - If the native memory tool is unavailable after diff approval, retain the approved before/after diff, mark the native write not_observed, and hand back that diff without editing MEMORY.md or USER.md directly.
+ 54 - If a native write returns staged: true with a pending_id, retain the approved diff and pending_id, mark saving not_observed, and hand back /memory approve <id> for the person without repeating the write.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

- 50 - If routing signals conflict, show the compact picker or ask one clarifying question.
- 51 - If wrapper metadata is unavailable, keep the recommendation advisory and avoid runtime claims.
+ 50 - If bounded recommend queries cannot run through CLI, plugin, or wrapper, retain only the imperative task, mark catalog selection not_observed, and hand back the workflow picker instead of a memorized recommendation.
+ 51 - If recommendation output contains only `omh-meta-router`, retain the consulted catalog evidence, mark workflow selection unresolved, and hand back the task remainder for a bounded re-query rather than selecting this skill.
+ 52 - If stage re-queries still leave a multi-stage task low-confidence, retain the consulted catalog evidence, mark the proposed chain tentative, and hand back a clarification question about the unresolved stage.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 65 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 65 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 60 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 60 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Local scripts → After Network or broad shell

d8697e46651e → 98284f305f6a

- 65 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 65 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Local scripts

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

+ 53 - Any print-and-rebuild probe carried the session's probe tag, and an observed search for that tag returns nothing.
- 112 - native_fault_statement/v1 separates the observed symptom from the assumed cause and names the reproduction command
+ 113 - native_fault_statement/v1 separates the observed symptom from the assumed cause, names the reproduction command, quotes the observed failure beside the reported one (`symptom_match`), and records the `minimised` reproduction

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 52 - If required context is missing, ask one blocking question or route back to the narrower workflow.
- 53 - If runtime or wrapper evidence is unavailable, keep the status as not_observed and expose the next observable action.
+ 52 - If source notes are missing, retain the cadence, audience, and time-window scaffold, mark minutes and decisions not_observed, and hand back operation_artifact/v1 with the explicit missing-notes boundary.
+ 53 - If notes leave an action item's owner or due date unstated, retain its source-linked outcome and dependencies, mark follow-up readiness unresolved, and hand back the action item history with unknown fields for confirmation.
+ 54 - If a retro proposal has no decision or owner acknowledgement, retain the source notes, mark acceptance unresolved, and hand back the proposal outside the decision log for explicit confirmation before an implementation handoff.
- 57 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 58 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 52 - If provider metrics are unavailable, report only local metadata and mark provider truth not_observed.
- 53 - If cost or latency looks risky, surface a warning plus the next measurement rather than a completion claim.
+ 52 - If the metric window is missing, keep supplied payloads, mark quality unresolved; hand back the next measure.
+ 53 - If a budget threshold is missing, keep the observed cost, mark advice tentative; hand back the gaps.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 51 - If required context is missing, ask one blocking question or route back to the narrower workflow.
- 52 - If runtime or wrapper evidence is unavailable, keep the status as not_observed and expose the next observable action.
+ 51 - If status evidence covers only part of the requested time window, retain the dated observations in the status summary, mark uncovered status not_observed, and hand back the missing periods and affected priorities.
+ 52 - If a blocker needs a code fix but no follow-up is accepted, retain the observed blocker and risk, mark remediation unresolved, and hand back the follow-up action for acceptance before delegation.
- 56 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 56 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 64 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 64 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 53 - If required context is missing, ask one blocking question or route back to the narrower workflow.
- 54 - If runtime or wrapper evidence is unavailable, keep the status as not_observed and expose the next observable action.
+ 53 - If interview evidence cannot support a candidate debrief, retain the role outcomes, must-have versus trainable criteria, and scorecard, mark the recommendation unresolved, and hand back the missing evidence to the decision owner.
+ 54 - If policy constraints need a jurisdiction-specific ruling, retain the role brief and interview scorecard, mark the policy conclusion unresolved, and hand back the flagged questions to `omh-legal-compliance-review`.
- 58 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 58 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 47 - When the plan describes code changes, it has a `Review Focus` section with at most 5 uncovered inputs and their owning tasks, or one stating the check found none.
+ 47 - When the plan describes code changes, it has a `Review Focus` section with at most 5 uncovered inputs and their owning task ids, or one stating the check found none.
+ 48 - When the plan describes code changes, every task has an id, its dependencies, the files it may write, and a done check, and the last task is the closing gate.
+ 49 - When the plan describes code changes, each code task is a vertical slice whose done check exercises its behavior end to end, or a prefactor ordered before the slices, or a wide change handed to `refactor-plan`, or the one task that writes a file every slice only appends to; the closing gate and fix tasks aside.
- 52 - If acceptance criteria or verification are missing, route back to clarification before handoff.
- 53 - If assumptions materially affect the plan, keep them visible and avoid treating the plan as accepted.
+ 54 - If no acceptance answer comes, keep the draft unaccepted, hand back the question, prepare no handoff.
+ 55 - If a verification step has no outside judge, mark it unresolved and hand back the missing judge.
- 81 - When the plan describes code changes, add a `Review Focus` section listing at most 5 inputs or failure modes the requirements imply that no planned test covers, likeliest first, each with its expected behavior and the owning task that will add the test; an input the requirements do not mention must still leave the result intact, and when the check turns up nothing the section says it found none.
+ 83 - When the plan describes code changes, add a `Review Focus` section listing at most 5 inputs or failure modes the requirements imply that no planned test covers, likeliest first, each with its expected behavior and the id of the owning task that will add the test; an input the requirements do not mention must still leave the result intact, and when the check turns up nothing the section says it found none.
+ 85 - When the plan describes code changes, give each task a stable id (`T1`, `T2`, ...), the ids it depends on, the files it may write, and its done check (a checkable step as above; for a code task, a command with the result that counts as passing - the test's run command, a gate, or a grep for the signature - while a task that is not code may keep an observable check); when two tasks write the same file, one lists the other directly among its dependencies, since an order through a third task does not count. These carry a `dag-ulw` node's id, depends_on and write_scope, and the done check feeds its VERIFY and STOP WHEN; converting still adds each node's label, category and full prompt, and a baseline node before a full-suite check.
+ 86 - When the plan describes code changes, cut the code tasks as vertical slices: each delivers one behavior through every layer it needs, and its done check drives that behavior through the surface a user or caller reaches, so an integration failure shows in the task that caused it rather than first in the closing gate. Do not split by layer to keep write scopes apart: slices that write the same file are ordered, each later slice listing every earlier slice that writes it directly among its dependencies, since an order through a slice between them does not count; a file every slice only appends to (a changelog, a generated index) may instead be written by one task after the slices; when that file is generated and drift-gated (a generated index), each slice's done check excludes that file's gate and the later task runs it. A layer-only task is either a prefactor that makes the later slices easy and comes before them, or a wide mechanical change across many callers, handed to `refactor-plan` (the new form beside the old, callers moved in groups while both stay green, the old form deleted last) instead of a split by layer.
+ 87 - When the plan describes code changes, its last task is the closing gate, named as the verification task (in `dag-ulw`, a node with `role: verify`): it depends on every other task, runs the real verification commands with the result that counts as passing, is run by someone other than the agent that built the work, and ends with a review. The plan states that a review that is not clean adds a fix task with its own id and done check and then reruns the gate instead of ending the plan - at most two fix rounds, or fewer when the engine running the plan stops sooner, after which the open findings are reported.
- 88 - Plan acceptance approves the plan content, not execution: after acceptance, recommend the follow-on path that fits the work's shape — `ultrawork` durable checkpoints for progress that must survive sessions as a checkpointed ledger, `ultrawork` coordinated lanes for an accepted plan split into disjoint parallel lanes, `ultrawork` single-owner persistence for one already-scoped task with a single owner, `ultrawork` for one bounded delivery cycle, or a direct selected executor/runtime handoff for a single prepared coding change — state the fit reason in one line, and start it only after the user's explicit go-ahead.
+ 93 - Plan acceptance approves the plan content, not execution: after acceptance, recommend the follow-on path that fits the work's shape — `dag-ulw` for an accepted plan whose tasks form a graph (several tasks with disjoint write scopes or an order between them, handing work on through files, plus a separate verification task), `ultrawork` durable checkpoints for progress that must survive sessions as a checkpointed ledger, `ultrawork` coordinated lanes for an accepted plan split into disjoint parallel lanes, `ultrawork` single-owner persistence for one already-scoped task with a single owner, `ultrawork` for one bounded delivery cycle, or a direct selected executor/runtime handoff for a single prepared coding change — state the fit reason in one line, and start it only after the user's explicit go-ahead.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 55 - If acceptance criteria or verification are missing, route back to clarification before handoff.
- 56 - If assumptions materially affect the plan, keep them visible and avoid treating the plan as accepted.
+ 55 - If the problem gate has not returned validated, retain supplied product evidence as unvalidated, mark PRD readiness unresolved, and hand back the problem and user context to `omh-product-discovery-validation`.
+ 56 - If the decision owner has not resolved prioritization tradeoffs, retain the drafted PRD and roadmap options, mark roadmap commitment unresolved, and hand back the tradeoffs and open questions for that owner.
+ 57 - If Jira or Linear updates lack an explicit target and approval, retain the prepared PRD, mark roadmap-system mutation not_observed, and hand back the target and approval questions for `omh-apps`.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 52 - If the reviewed target is missing, inspect the requested artifact or ask one target question.
- 53 - If independent verification is unavailable, report the gap and avoid an approval-style claim.
+ 52 - If build or CI evidence is missing for the target release, retain the evidenced readiness_matrix/v1 rows, mark readiness unresolved, and hand back release_gate_verdict/v1 on HOLD with the missing evidence list.
+ 53 - If rollback owner, health thresholds, or recovery path is missing, retain the evidenced readiness_matrix/v1 rows, mark rollback readiness unresolved, and hand back rollback_and_monitoring_plan/v1 with the exact gaps.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 51 - If required context is missing, ask one blocking question or route back to the narrower workflow.
- 52 - If runtime or wrapper evidence is unavailable, keep the status as not_observed and expose the next observable action.
+ 51 - If provider ID or profile ID is missing, retain the supplied capability requirements, mark provider_profile_posture/v1 unresolved, and hand back a request for the missing ID without reading any secret.
+ 52 - If secret-presence metadata is absent, retain the named provider, profile, and requested capabilities, mark secret presence not_observed, and hand back metadata-only secret requirements plus allowed and prohibited actions.
+ 53 - If only a host observation reference is supplied, retain it as metadata, mark provider connectivity not_observed, and hand back provider_profile_posture/v1 without claiming credential validation or making a provider call.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

- 57 - If reconnaissance finds the direction itself is unsettled, route back to `ralplan` before ordering phases.
+ 57 - If reconnaissance finds the direction itself is unsettled, route back to `ulw-plan` before ordering phases.
+ 58 - If the module to deepen is chosen but its interface is not, shape it under contrasting constraints and recommend one design before phase 1; that is not a contested direction.
- 86 - Stop at the approval gate and hand the user the go/no-go, whole plan or first phase.
+ 87 - Stop at the approval gate and hand the user the go/no-go, whole plan or first phase; when the approved phases split into groups with disjoint files plus a verification step, a whole-plan go may also name `dag-ulw` with the fit reason in one line, started only on the user's explicit go-ahead: the phase order becomes `depends_on` and each phase's verification command its node's VERIFY. `dag-ulw` runs without pausing between phases and neither commits nor rolls back, so say the per-phase commits are cut after the run and the rollback points stay with the user, or take the first-phase path.
- 94 - the decided target shape (what moves where), or a pointer to the accepted plan that decided it
+ 95 - the decided target shape (what moves where), or a pointer to the accepted plan that decided it; for a chosen deepening candidate whose interface is still open, shape it first per `omh-refactor-plan/references/interface-shaping.md`

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

+ 89 - For a read-modify-write, an upsert, a queue table, or a deadlock, use the concurrency section of the lock tables: name the isolation level, the row lock and its order, and which error codes retry the whole transaction.
+ 90 - Give non-transactional DDL such as `CREATE INDEX CONCURRENTLY` its own migration marked non-transactional; never drop `CONCURRENTLY` to satisfy the migration tool.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 53 - If required context is missing, ask one blocking question or route back to the narrower workflow.
- 54 - If runtime or wrapper evidence is unavailable, keep the status as not_observed and expose the next observable action.
+ 53 - If metrics do not cover the named service and time window, retain the cited incident notes, mark SLO and error-budget conclusions unresolved, and hand back the evidence and missing-evidence list with the uncovered window.
+ 54 - If remediation closure has no supporting reference, retain the source-backed reliability review, mark closure not_observed, and hand back the remediation follow-ups and missing closure evidence without claiming the fix completed.
- 58 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 58 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 52 - If input evidence is incomplete, mark the section as pending rather than fabricating a report claim.
- 53 - If delivery or attachment is unavailable, keep the report package prepared_not_observed.
+ 52 - If no presentation tool can produce a binary deck, retain the PPT-ready Markdown or JSON outline, mark binary export not_observed, and hand back the outline, assumptions, and missing-input list for separate deck generation.
+ 53 - If the tool returns "rendered": false (it still exits 0) or its image count differs from slide_count, keep the deck's outline read-back, leave render QA not observed, and hand back the missing-tool or image-count gap for that file.
+ 54 - If the requested achievements export fails, retain the rest of the report package, mark the badge section not_observed, and hand back the missing output from omh achievements export --format md instead of inventing badges.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 66 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 66 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 53 - If sources cannot be accessed, state the retrieval gap and use only observed local context.
- 54 - If evidence is thin or one-sided, lower confidence and ask for a narrower source boundary.
+ 53 - If ledger verification cites zero distinct sources, retain the source ledger and evidence table, mark the citation check not_observed, and hand back the draft with the missing [n] source IDs for correction.
+ 54 - If an evidence row's quote cannot be attached from the page text, retain the retrieved source and other supported rows, mark that claim unresolved, and hand back the unmatched wording and source for correction.
+ 55 - If a pricing comparison mixes periods, currencies, or denominators, retain its dated source rows, mark the comparison tentative, and hand back the missing common basis before drawing a decision implication.
- 58 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 59 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.
- 82 - When Hermes' bundled `grounded-citations` skill is available, load it and register each source in its ledger when it is retrieved (`scripts/sources.py add`), and carry two more fields on every evidence row: the ledger id written as `[n]` - a bare number in a cell is not a citation to `verify`, so that row would go unchecked - and the verbatim quote attached with `sources.py quote <id> --text <quote> --from <page text file>`, which rejects wording that file does not contain. Keep the run's ledger in a file of its own, passed as `--ledger <path>` before the subcommand on every call: the default ledger is shared by the whole profile, and another run's reset rebinds its ids. A row whose quote could not be attached says so, and a claim that rests on model knowledge is marked `[unverified]` and listed as unresolved, never entered as evidence. Run `sources.py verify <draft> --evidence`, since every evidence row is meant to carry its quote, and report the citation check from that run, or as `not_observed`: the exit status of `sources.py verify <draft>` is the observed check - `0` is a pass and any other status is not - and `verify --evidence` also fails a draft whose cited source carries no verbatim quote. An exit `0` whose `info: stats:` line reads `0 distinct source(s) cited` checked nothing and is reported as `not_observed`; it is a citation check, not fact verification. Without the skill, keep the source and its quoted wording in the row and report the citation check as `not_observed`.
+ 83 - When Hermes' bundled `grounded-citations` skill is available, load it and register each source in its ledger when it is retrieved (`scripts/sources.py add`), and carry two more fields on every evidence row: the ledger id written as `[n]` - a bare number in a cell is not a citation to `verify`, so that row would go unchecked - and the verbatim quote attached with `sources.py quote <id> --text <quote> --from <page text file>`, which rejects wording that file does not contain. Keep the run's ledger in a file of its own, passed as `--ledger <path>` before the subcommand on every call: the default ledger is shared by the whole profile, and another run's reset rebinds its ids. A row whose quote could not be attached says so, and a claim that rests on model knowledge is marked `[unverified]` and listed as unresolved, never entered as evidence. Run `sources.py verify <draft> --evidence`, since every evidence row is meant to carry its quote, and report the citation check from that run, or as `not_observed`: the exit status of `sources.py verify <draft>` is the observed check - `0` is a pass and any other status is not - and `verify --evidence` also fails a draft whose cited source carries no verbatim quote. An exit `0` whose `info: stats:` line reads `0 distinct source(s) cited` checked nothing and is reported as `not_observed`; it is a citation check, not fact verification. After that verify, run `omh research receipts verify --citations <ledger> --draft <draft>` and report its fetched coverage from the run, or as `not_observed`; a receipt proves only that the page was opened in this session tree, never that it supports the claim. Without the skill, keep the source and its quoted wording in the row and report the citation check as `not_observed`.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 55 - If sources cannot be accessed, state the retrieval gap and use only observed local context.
- 56 - If evidence is thin or one-sided, lower confidence and ask for a narrower source boundary.
+ 55 - If one watch source cannot be read, retain covered sources' findings and the failed source's previous cutoff, mark its coverage unresolved, and hand back that source and uncovered window rather than reporting a quiet period.
+ 56 - If the synthesis tool cannot process collected findings, retain raw findings in source_inbox/v1, mark synthesis not_observed, and hand back research_department_plan/v1 with the synthesis-tool readiness gap.
+ 57 - If schedule creation has no cronjob_manage result or cron listing, retain the cadence and Scout, Analyst, and Briefer plan, mark job creation not_observed, and hand back the prepared job description for separately approved scheduling.
- 60 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 61 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 59 - If routing signals conflict, show the compact picker or ask one clarifying question.
- 60 - If wrapper metadata is unavailable, keep the recommendation advisory and avoid runtime claims.
+ 59 - If a skill is not installed, keep the intent; hand back the install gap as unresolved.
+ 60 - If executor readiness is unknown, keep routing; hand back the check as not_observed.
- 68 - Intent -> plan: `deep-interview`, `ralplan`, `plan`, `loop`.
+ 68 - Intent -> plan: `deep-interview`, `ralplan`, `plan`, `dag-ulw`, `loop`.
- 81 - Never let that autonomy bypass existing confirmation gates for destructive changes, credentials, external writes, deployment, executor dispatch, or starting a follow-on workflow engine (`ultrawork` — including its coordinated-scope, single-owner-persistence, delivery-boundary, and durable-checkpoint capabilities — `loop`, `ultraqa`) from another skill's output: an accepted plan or clarified brief is planning evidence, not permission — recommend the engine that fits the work's shape and wait for the user's explicit go-ahead. Do not claim that a native skill, subagent, review, CI, or merge ran unless matching observation exists.
+ 81 - Never let that autonomy bypass existing confirmation gates for destructive changes, credentials, external writes, deployment, executor dispatch, or starting a follow-on workflow engine (`dag-ulw`, `ultrawork` — including its coordinated-scope, single-owner-persistence, delivery-boundary, and durable-checkpoint capabilities — `loop`, `ultraqa`) from another skill's output: an accepted plan or clarified brief is planning evidence, not permission — recommend the engine that fits the work's shape and wait for the user's explicit go-ahead. Do not claim that a native skill, subagent, review, CI, or merge ran unless matching observation exists.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 47 - The durable fact, source evidence, retrieval hint, and staleness risk are recorded.
- 48 - Uncertain or conflicting knowledge is marked as review-needed rather than permanent truth.
- 49 - Separate coding or docs tasks are extracted instead of buried in notes.
+ 47 - Every candidate in principle_candidate_set/v1 cites the repeated observations it rests on; a candidate resting on one observation stays needs-evidence in review_queue/v1.
+ 48 - Every candidate is classed mechanical or judgement before it is worded. A mechanical candidate is a miss a machine can find from the text alone (a forbidden call, a disallowed import, a file in the wrong place), and its destination is a check (a linter rule, test, hook, or CI step); when the repository already has a check for it that never runs or no longer fails, the candidate is repairing that check, not adding another. Its approved_patch_handoff/v1 names the check and where it is wired, and sets as an acceptance criterion the run the implementer must observe: the check fails with the violation injected and passes once it is reverted. A mechanical candidate outside the chosen destination boundary is reported as a check proposal outside the boundary, never reworded as prose. Only judgement candidates, which need a reader (consistency across files, fit with the code around them), are worded as prose.
+ 49 - duplication_conflict_report/v1 names the existing rule or check each candidate repeats or contradicts, or says none was found and where it looked.
+ 50 - Nothing reached a destination without an approval recorded in review_queue/v1.
- 53 - If source evidence conflicts, route to memory or knowledge review before writing durable guidance.
- 54 - If the fact may be stale, record the staleness warning and next refresh action.
+ 54 - If a principle rests on one observation, retain its source reference in principle_candidate_set/v1, mark the rule tentative and needs-evidence in review_queue/v1, and hand back the candidate for repeated observations.
+ 55 - If approval is missing, retain principle_candidate_set/v1 and duplication_conflict_report/v1, mark implementation not_observed, and hand back review_queue/v1 without producing approved_patch_handoff/v1.
+ 56 - If skill_manage stages an approved change, retain the approved candidate and staged result, mark the skill update not_observed, and hand back the pending ID for the person's `/skills approve <id>`.
- 91 - destination boundary: AGENTS, skill catalog, prompt, docs, memory, or no-write review
+ 93 - destination boundary: AGENTS, skills, prompt, docs, memory, check, or no-write review
- 106 - principle_candidate_set/v1 with source references, repeated pattern, candidate wording, scope, non-goals, and risk
+ 108 - principle_candidate_set/v1 with source references, repeated pattern, mechanical or judgement class, candidate wording or check, scope, non-goals, and risk

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 52 - If provider metrics are unavailable, report only local metadata and mark provider truth not_observed.
- 53 - If cost or latency looks risky, surface a warning plus the next measurement rather than a completion claim.
+ 52 - If the named Hermes session has no row or recorded tokens, retain supplied timings and context data in run_efficiency_report/v1, mark host_usage not_observed, and hand back the session ID for a recorded-usage check.
+ 53 - If session_activity_receipt/v1 omits tool durations or context-budget metrics, retain its available observations, mark the missing metrics not_observed rather than zero, and hand back the exact missing measurements.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 50 - The gate list names the exact commands the executor must run and pass.
+ 50 - The gate list names the exact commands the executor must run and pass, covering every target, the feature extremes, doc tests, and the declared MSRV.
- 82 - Load `references/rust-discipline.md` for the ownership, error, and API rules, and name the gate commands from it rather than assuming `cargo build` is the whole bar.
+ 82 - Load `references/rust-discipline.md` for the ownership, error, API, binary-interface, and dependency rules, and name the gate commands from it rather than assuming `cargo build` is the whole bar.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 54 - If evidence is mostly assumption, label it and recommend a research or feedback-triage pass.
- 55 - If the decision owner is missing, keep the output as options rather than accepted strategy.
+ 54 - If consent or suppression status is unknown, retain sales_opportunity_evidence_record/v1 and sales_qualification_state/v1, mark sequence eligibility unresolved with HOLD, and hand back the missing eligibility questions without drafting outreach.
+ 55 - If no observed buyer response supports a commitment, retain the eligible sales_draft_sequence/v1, mark confirmation not_observed, and hand back sales_handoff_disposition/v1 with proposed actions, owners, timing, and exit criteria.
- 59 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 59 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Instructions only

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 54 - If the reviewed target is missing, inspect the requested artifact or ask one target question.
- 55 - If independent verification is unavailable, report the gap and avoid an approval-style claim.
+ 54 - If verify-new lacks an observed credential check, retain the redacted risk matrix and credential_rotation_sequence/v1, mark replacement unresolved, and hand back the verify-new evidence requirement before the operator revokes the old credential.
+ 55 - If no single local plugin directory is explicitly named, no plugin audit finding stands; mark plugin_risk_audit/v1 not_observed and hand back the directory question without importing, registering, or activating any plugin.
+ 56 - If a proposed deny glob also blocks commands the workflow needs, retain safe_action_policy/v1, mark proposed_approval_deny_list/v1 unresolved, and hand back the conflicting globs and needed commands for user revision without changing config.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Network or broad shell → After Network or broad shell

d8697e46651e → 98284f305f6a

+ 30 - If the selected concept is a 60-120 s launch film of a product the team can run and capture, read `references/launch-film.md` before the act sheet and scene authoring; a promo of 15 s or less keeps the one-scene rule.
+ 41 - A promo or social clip of 15 s or less holds one hero scene with no meta text, one number smaller than the logo, and the official logo unboxed at 25% or more of the frame height; check a still sheet against `references/craft-design.md` (Short promo clips) before rendering.

Compare on GitHub

Before Not checked yet → After Network or broad shell

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 62 - If the user asks for facts or citations, route to `research`.
+ 62 - If the user asks for facts or citations, route to `ulw-research`.
- 68 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 68 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Local scripts → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Local scripts

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 54 - If feedback lacks source or severity, ask for the missing signal before coding handoff.
- 55 - If the item is actually a plan or research request, route to that workflow instead of triage.
+ 54 - If customer impact or reproduction evidence is missing, retain the stated case facts, mark severity tentative, and hand back the reply draft and missing repro, account, or entitlement evidence list.
+ 55 - If no escalation owner or path is available, retain the issue/severity/impact/escalation matrix, mark ownership unresolved, and hand back the internal next-step and owner handoff brief with the owner question.
- 59 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 59 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → 98284f305f6a

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 3 description: "[omh] Accumulated tech debt to rank: build the severity-by-effort debt ledger from observed repo evidence - orient, audit the named dimensions with file:line citations, rank fixes and quick wins - and reconcile RESOLVED/NEW/CARRIED against the previous ledger on rerun. Use when the user says: tech-debt-audit, tech debt, tech debt audit, technical debt, technical debt audit, tech debt ledger, debt ledger, audit our tech debt."
+ 3 description: "[omh] Accumulated tech debt or architecture deepening to rank: build the severity-by-effort debt ledger from observed repo evidence - orient, audit the named dimensions with file:line citations, rank fixes and quick wins - and reconcile RESOLVED/NEW/CARRIED on rerun. Use when the user says: tech-debt-audit, tech debt, tech debt audit, technical debt, technical debt audit, tech debt ledger, debt ledger, audit our tech debt."
- 68     Strong routing signals: `tech-debt-audit`, `tech debt`, `tech debt audit`, `technical debt`, `technical debt audit`, `tech debt ledger`, `debt ledger`, `audit our tech debt`, `tech debt report`, `code debt audit`, `where is our tech debt`, `기술부채`, `기술 부채`, `기술부채 감사`, `기술부채 감사해줘`, `기술부채 점검`, `기술부채 장부`, `부채 원장`, `техдолг`, `технический долг`, `аудит техдолга`, `аудит технического долга`
+ 68     Strong routing signals: `tech-debt-audit`, `tech debt`, `tech debt audit`, `technical debt`, `technical debt audit`, `tech debt ledger`, `debt ledger`, `audit our tech debt`, `tech debt report`, `code debt audit`, `where is our tech debt`, `improve the codebase architecture`, `improve the architecture of this repo`, `deepening opportunities in this codebase`, `shallow modules in this codebase`, `기술부채`, `기술 부채`, `기술부채 감사`, `기술부채 감사해줘`, `기술부채 점검`, `기술부채 장부`, `부채 원장`, `코드베이스 아키텍처 개선해줘`, `코드 아키텍처 개선해줘`, `техдолг`, `технический долг`, `аудит техдолга`, `аудит технического долга`
+ 84 - For an architecture-improvement or deepening request, load `omh-tech-debt-audit/references/architecture-deepening.md`: candidates are architectural-decay ledger rows proven shallow by the deletion test, never by size, and the audit stops at ranked candidates with one named top pick - it proposes no interface.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Network or broad shell → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Network or broad shell

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 52 - If required context is missing, ask one blocking question or route back to the narrower workflow.
- 53 - If runtime or wrapper evidence is unavailable, keep the status as not_observed and expose the next observable action.
+ 52 - If tool_search finds a deferred native tool but access is untested, retain its present classification, mark API access not_observed, and hand back the tool_call step for a user-approved check instead of listing the tool as missing.
+ 53 - If a required MCP server or CLI is absent, retain the present, optional, and unsafe inventory entries, mark workflow readiness unresolved, and hand back the missing dependency and installation owner without installing it.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 52 - If the expected behavior is unclear, route back to plan before running adversarial checks.
- 53 - If verification fails, return to fix or research with the failed signal instead of advancing.
+ 52 - If a required check fails with an unknown origin, keep the other fresh results, mark the claim unresolved on HOLD, and hand back the same check rerun at the merge base as the missing evidence.
+ 53 - If a fix test has only a fresh green result, retain that result within its scope, mark the claim unresolved on HOLD with regression_witness_missing, and hand back the required pre-fix red witness and its source.
+ 54 - If a declared generated path has no known regeneration command, retain verification_matrix/v1, mark generated-output verification unresolved, and hand back generated_artifact_provenance/v1 with the command gap.
+ 85 - A fix claim whose evidence is a new or changed test needs a regression witness: an observed red of that test, for the bug's reason, on the pre-fix tree (the fix commit's parent, or the merge-base when the fix is the whole change), recorded with its source -- a run in a clean detached worktree at that tree, a CI job on a commit that adds only the test, or the debugging session's run before the fix. A collection or import error there is not a witness. For an intermittent fault the witness is the failure rate over N runs before beside 0/N at the same N after. When the test calls a seam the fix added, force the red by reverting only the behavioural lines, keep the seam, and show that diff. When the fix changes only test code, the witness is the old test's measured failure rate on the pre-fix tree beside the new test's 0/N at the same N. A fix test only ever observed green is HOLD with `regression_witness_missing`. A fix recorded with `regression_seam: none_correct` and its reason has no fix test; its evidence is the reproduction loop observed red before and green after (for an intermittent fault, its rate over N runs before beside 0/N after), and the test witness rule does not apply.
- 112 - claim_verdict/v1 with PASS, HOLD, or BLOCK and exact missing or failed checks, each failed one with its origin
+ 114 - claim_verdict/v1 with PASS, HOLD, or BLOCK, a HOLD reason of `regression_witness_missing` when a fix test has no observed pre-fix red, and exact missing or failed checks, each failed one with its origin

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 51 - If transcript confidence or intent is weak, ask one short clarification before action.
- 52 - If platform delivery is unavailable, keep the response in chat and mark delivery not_observed.
+ 51 - If a dictated request names no target context, nothing is actionable yet; mark the next action unresolved and hand back one target question before preparing voice-operator/v1.
+ 52 - If a clarified voice request still lacks confirmation for a risky action, retain the voice-operator/v1 guidance, mark execution not_observed, and hand back the proposed next action for explicit approval.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 66 - If leads keep expanding past one round, hand the question to `research` with the sources already gathered.
+ 66 - If leads keep expanding past one round, hand the question to `ulw-research` with the sources already gathered.
- 70 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 70 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.
- 95 - When that ledger is in use, report the citation check from a run, or as `not_observed`: the exit status of `sources.py verify <draft>` is the observed check - `0` is a pass and any other status is not - and `verify --evidence` also fails a draft whose cited source carries no verbatim quote. An exit `0` whose `info: stats:` line reads `0 distinct source(s) cited` checked nothing and is reported as `not_observed`, and so is an answer given only in chat: `verify` reads a draft file and exits `2` without one, so write the answer to a file first when the check matters. The run compares cited ids and the Sources block with the ledger; it does not test whether a claim is true, so a passing citation check is never reported as fact verification. Without the ledger the citation check is `not_observed`.
+ 95 - When that ledger is in use, report the citation check from a run, or as `not_observed`: the exit status of `sources.py verify <draft>` is the observed check - `0` is a pass and any other status is not - and `verify --evidence` also fails a draft whose cited source carries no verbatim quote. An exit `0` whose `info: stats:` line reads `0 distinct source(s) cited` checked nothing and is reported as `not_observed`, and so is an answer given only in chat: `verify` reads a draft file and exits `2` without one, so write the answer to a file first when the check matters. The run compares cited ids and the Sources block with the ledger; it does not test whether a claim is true, so a passing citation check is never reported as fact verification. After that verify, run `omh research receipts verify --citations <ledger> --draft <draft>` and report its fetched coverage from the run, or as `not_observed`; a receipt proves only that the page was opened in this session tree, never that it supports the claim. Without the ledger the citation check is `not_observed`.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 52 - If required context is missing, ask one blocking question or route back to the narrower workflow.
- 53 - If runtime or wrapper evidence is unavailable, keep the status as not_observed and expose the next observable action.
+ 52 - If an in-scope plugin, MCP, or runtime surface cannot be inspected, retain the observed surface_inventory/v1 entries, mark that surface not_observed, and hand back the evidence gap rather than call it a missing capability.
+ 53 - If config risks cannot be checked without exposing secret values, retain redacted key names and paths, mark those checks unresolved, and hand back config_security_findings/v1 with risk categories for later remediation.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 61 - If the user moves from terminology to implementation, summarize confirmed understanding and hand off to `ralplan`, `ulw-plan`, or the selected coding owner only after a separate go-ahead.
+ 61 - If the user moves from terminology to implementation, summarize confirmed understanding and hand off to `ulw-plan` or the selected coding owner only after a separate go-ahead.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 105 number (a reply by number would collide with it, in the list below and in a messaging picker).
+ 105 number (a reply by number would collide with it). If memory suggests candidates, draw one
+ 106 from the request or repository, so memory never fills every slot.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

+ 112 - When the stop ladder names `replan_stuck_loop_once`, re-plan once from the goal ledger and recorded failures alone, not the transcript or recalled memory; record the new approach and continue; a second stop names the blocker step.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 58 - If the selected executor is unavailable, ask for Codex, Claude Code, Hermes, or another runtime before retrying.
- 59 - If dispatch or result evidence is missing, keep the handoff prepared_not_observed and expose the next observable action.
+ 58 - If this run lacks an explicit coding-owner choice, no handoff stands yet; mark prompt composition unresolved and hand back the choose_executor question without selecting a CLI.
+ 59 - If the scratch-edit permission probe fails, retain the composed prompt and discovered skill names, mark execution not_observed, and hand back the denied permission and required grant change before retrying.
+ 60 - If dispatch returns no session or thread id, retain the dispatch receipt, mark steering unresolved, and hand back the explicit unsteerable note and any observed unit result without claiming a resumable session.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 67 - If current-source evidence is missing, route a `research` step before accepting the plan.
- 68 - If the user asks for implementation after acceptance, recommend the follow-on path that fits the work's shape (`ultrawork` with the matching capability — durable checkpoint, coordinated lanes, single-owner persistence, or one delivery cycle — or a direct selected executor handoff) with a one-line fit reason, and start it only on the user's explicit go-ahead — never auto-start an engine from acceptance alone.
+ 67 - If current-source evidence is missing, route a `ulw-research` step before accepting the plan.
+ 68 - If the user asks for implementation after acceptance, recommend the follow-on path that fits the work's shape (`dag-ulw` when the accepted lanes form a graph with a verification lane, `ulw-work` with the matching capability — durable checkpoint, coordinated lanes, single-owner persistence, or one delivery cycle — or a direct selected executor handoff) with a one-line fit reason, and start it only on the user's explicit go-ahead — never auto-start an engine from acceptance alone.
- 95 - List every lane of the accepted plan in node-prompt shape - `TASK`, `DELIVERABLE`, `SCOPE`, `VERIFY`, `STOP WHEN` - with `depends_on` per lane, so `ultrawork` can prepare board rows from the plan without re-planning; planning itself stays a bounded in-session lane.
+ 95 - List every lane of the accepted plan in node-prompt shape - `TASK`, `DELIVERABLE`, `SCOPE`, `VERIFY`, `STOP WHEN` - with `depends_on` per lane, so `dag-ulw` nodes or `ultrawork` board rows come from the plan without re-planning; planning itself stays a bounded in-session lane.
- 110 - Plan acceptance approves the plan content, not execution: after acceptance, recommend the follow-on path that fits the work's shape — `ultrawork` durable checkpoints for progress that must survive sessions as a checkpointed ledger, `ultrawork` coordinated lanes for an accepted plan split into disjoint parallel lanes, `ultrawork` single-owner persistence for one already-scoped task with a single owner, `ultrawork` for one bounded delivery cycle, or a direct selected executor/runtime handoff for a single prepared coding change — state the fit reason in one line, and start it only after the user's explicit go-ahead.
+ 110 - Plan acceptance approves the plan content, not execution: after acceptance, recommend the follow-on path that fits the work's shape — `dag-ulw` for an accepted plan whose tasks form a graph (several tasks with disjoint write scopes or an order between them, handing work on through files, plus a separate verification task), `ultrawork` durable checkpoints for progress that must survive sessions as a checkpointed ledger, `ultrawork` coordinated lanes for an accepted plan split into disjoint parallel lanes, `ultrawork` single-owner persistence for one already-scoped task with a single owner, `ultrawork` for one bounded delivery cycle, or a direct selected executor/runtime handoff for a single prepared coding change — state the fit reason in one line, and start it only after the user's explicit go-ahead.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 52 - If the expected behavior is unclear, route back to plan before running adversarial checks.
- 53 - If verification fails, return to fix or research with the failed signal instead of advancing.
+ 52 - If a scenario needs a real install or external service the harness cannot reach, retain other observed pass/fail evidence, mark that scenario not_observed, and hand back its manual test guide with setup, do, expect, and broken signals.
+ 53 - If a hostile browser scenario would submit, buy, send, or delete on a live target without approval, retain read-only scenario results, mark that action not_observed, and hand back the target and exact action for user approval.
+ 54 - If a failing browser scenario has no screenshot or console readback, keep its observed failure, mark diagnosis tentative, and hand back the scenario, target, and missing capture for a bounded rerun.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

- 72 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+27 more`) - research, signals, ops, and briefings.
+ 72 - Current lane: **Research and company ops** (`product-docs`, `source-finder`, `web-research`, `research`, `model-optimization`, `inference-serving`, `model-finetuning`, `research-brief`, `+28 more`) - research, signals, ops, and briefings.
- 100 - When Hermes' bundled `grounded-citations` skill is available, load it and register each source in its ledger when it is retrieved (`scripts/sources.py add`). Keep the run's ledger in a file of its own, passed as `--ledger <path>` before the subcommand on every call: the default ledger is shared by the whole profile, and another run's reset rebinds its ids. Use parent-only serialized writes: lanes return observed source entries and quote refs; the parent registers URLs, attaches quotes, and remaps lane ids before drafting. Lanes never write the shared ledger. When a human briefing is written, report the citation check from a run, or as `not_observed`: the exit status of `sources.py verify <draft>` is the observed check - `0` is a pass and any other status is not - and `verify --evidence` also fails a draft whose cited source carries no verbatim quote. An exit `0` whose `info: stats:` line reads `0 distinct source(s) cited` checked nothing and is reported as `not_observed`. It is a citation check, not fact verification, and it never stands in for the contested-claim gate. Without the skill, keep each source's URL and retrieval date in the research notes and report the citation check as `not_observed`.
+ 100 - When Hermes' bundled `grounded-citations` skill is available, load it and register each source in its ledger when it is retrieved (`scripts/sources.py add`). Keep the run's ledger in a file of its own, passed as `--ledger <path>` before the subcommand on every call: the default ledger is shared by the whole profile, and another run's reset rebinds its ids. Use parent-only serialized writes: lanes return observed source entries and quote refs; the parent registers URLs, attaches quotes, and remaps lane ids before drafting. Lanes never write the shared ledger. When a human briefing is written, report the citation check from a run, or as `not_observed`: the exit status of `sources.py verify <draft>` is the observed check - `0` is a pass and any other status is not - and `verify --evidence` also fails a draft whose cited source carries no verbatim quote. An exit `0` whose `info: stats:` line reads `0 distinct source(s) cited` checked nothing and is reported as `not_observed`. It is a citation check, not fact verification, and it never stands in for the contested-claim gate. After that verify, run `omh research receipts verify --citations <ledger> --draft <draft>` and report its fetched coverage from the run, or as `not_observed`; a receipt proves only that the page was opened in this session tree, never that it supports the claim. Without the skill, keep each source's URL and retrieval date in the research notes and report the citation check as `not_observed`.
- 103 - Decompose the question into orthogonal research axes and disambiguate named entities before any deep reading.
+ 103 - Decompose the question into orthogonal research axes and disambiguate named entities before any deep reading. Before answering from a source, confirm it studies the exact entity the question names; a different name is a different entity unless a source says they are the same, so a source on a related one is a lead, not an answer.

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub

Before Instructions only → After Network or broad shell

d8697e46651e → 98284f305f6a

Compare on GitHub

Before Not checked yet → After Instructions only

First observation → d8697e46651e

No earlier version was observed for comparison.

Compare on GitHub