Menu

Individual skill

Emit policy receipts for hard-rule agent skills with Pluribus

0upvotes

Uses the Pluribus skill-policy receipt recipe to make hard agent-skill rules auditable before and after writes. Best for Claude Code, OpenClaw, Cursor, Codex, and other coding-agent workflows where natural-language instructions such as forbidden services, generated-file boundaries, or preview-before-apply rules need explicit allow/refuse evidence.

skills/emit-policy-receipts-for-hard-rule-agent-skills-with-pluribus/SKILL.md

Read pinned source

Source commit fc97377adefd53e861c1d456aa1392e4a4e022a2

What was checked

Matches the observed repository HEAD
Yes
Scanner rules are current
No
Repository scan complete
No
Skill scan complete
No
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.

Standard SKILL.md format has not been attested at this commit. No installation command is available; inspect the source.

Repository context

Repository stars (not skill ratings)
62
Repository last updated
Nov 9, 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

# Emit policy receipts for hard-rule agent skills with Pluribus

Use this skill when an agent skill, project rule, hook, or runbook contains a hard boundary that should not depend on prose alone. Examples include “do not call internal services from unit tests”, “do not edit generated files”, “preview migrations before apply”, “never write outside this workspace”, or “refuse before touching customer data”. The Pluribus recipe turns those rules into a small privacy-safe `skill.policy.v1` receipt that records intended targets, allow/refuse decisions, refusal reasons, post-write guard results, and `stopped_at` if the agent must halt.

The goal is not to store prompts, transcripts, raw code, secrets, stack traces, or tool output. The receipt should only capture boundary metadata that a reviewer or a later agent can inspect: which targets were considered, why each target was allowed or refused, whether a write began, and whether a simple guard passed after the write. This is useful for Claude Code, OpenClaw, Cursor, Codex, and other agent runtimes where Skills, hooks, `CLAUDE.md`, `AGENTS.md`, or policy files can be loaded dynamically but the run still needs auditable evidence that the policy was followed.

## Installation

### OpenClaw

```bash
clawhub install emit-policy-receipts-for-hard-rule-agent-skills-with-pluribus
```
Back to top