Menu

Individual skill

hubo

0upvotes

Implements and tests code while an independent reviewer challenges every change until all findings are reconciled. Use only when the user explicitly invokes, selects, or mentions Hubo for a programming task.

skills/hubo/SKILL.md

Read pinned source

Source commit 037c52cf36142ef6c63a812d9df57bc9bcae08ba

What was checked

Matches the observed repository HEAD
Yes
Scanner rules are current
No
Repository scan complete
No
Skill scan complete
Yes
Observed behavior
Not checked yet
Last checked
Nov 6, 2026

These are observations at the recorded commit, not a guarantee of safety. Review the source before running it.

Install this skill

Choose an agent in your Stack or copy a command below. Commands fetch the pinned commit and copy only this skill directory. Existing destinations are refused. Copying a command does not install anything.

An install command is unavailable: current rules, complete repository and skill scans, an eligible behavior level and a permitted license are required.

Repository context

Repository stars (not skill ratings)
27
Repository last updated
Jul 29, 2026
License
MIT

Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.

Files and excerpt

File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.

Browse files
Read excerpt

# Hubo

Run Hubo only after explicit user activation. Coordinate exactly two persistent top-level roles in the current conversation:

- a pragmatic work agent whose lineage alone may edit files; and
- an independent, read-only review agent that challenges the implementation and its evidence.

Read [the complete Hubo workflow](references/workflow.md) before creating either role. Select exactly one current host from [the supported host adapters](references/hosts.md); never mix host APIs.

The work agent implements the smallest complete change and reports its diff, decisions, and verification. The review agent inspects the actual code and returns `CLEAR` or numbered findings. Relay every finding back to the same work agent, which must respond `FIXED`, `PUSHBACK`, or `NEEDS_USER` with evidence. Resume the same reviewer to verify the response, then continue without an arbitrary round limit.

Stop only when the reviewer returns `CLEAR` with every finding closed or withdrawn, or when a genuine requirement or technical-direction decision needs the user. Mirror every completed work and review report into the conversation, then include the complete chronological exchange in the final response. Never store the exchange in a file.
Back to top