agentsclimarketplace
Sign in

Universal skill governance

Skill the-long-ride/skill-master/skills/universal-skill-governance

Use when creating, reviewing, or publishing AI-agent skills that must run consistently across local agents and web LLM chat interfaces.From its SKILL.md

Install
npx -y skills add the-long-ride/skill-master --skill universal-skill-governance

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.6 KB, 538 tokens by cl100k_base, as published. Nobody here has run it

Universal Skill Governance

Overview

This skill defines a portable quality bar for agent skills. Use it to keep skills discoverable, executable, and verifiable across runtimes.

When to Use

Use this skill when:

  • A new skill is being created for broad reuse.
  • A skill behaves differently across agent platforms.
  • A skill is unclear, too verbose, or difficult to verify.
  • You need a release gate before sharing skills.

Do not use this skill when:

  • The task is a one-off local note.
  • Requirements can be enforced by static tooling alone.

Inputs and Tools

Inputs:

  • Target SKILL.md
  • Optional supporting references
  • Workspace-level skill index

Tools:

  • File search (rg, file listing)
  • AI-guided checklist verification
  • Optional local shell script verification
  • Human checklist review

Workflow

  1. Validate trigger quality
  • Ensure frontmatter description starts with Use when.
  • Ensure description explains only when to use the skill.
  1. Validate execution quality
  • Confirm workflow has concrete, ordered steps.
  • Confirm failure modes include explicit recovery.
  • Confirm output format is deterministic.
  1. Validate portability
  • Ensure no platform-specific assumptions are required to understand the skill.
  • Move runtime-specific details into references.
  1. Validate verification
  • Run AI-guided checklist review and produce PASS/PARTIAL/FAIL report.
  • If needed, run local audit script for automation parity.
  • Compare outcomes and resolve mismatches.
  • Fix all critical FAIL findings before publishing.
  1. Publish readiness
  • Update skill index metadata.
  • Add release notes if behavior changed.

Failure Modes

  • Trigger drift: Description includes workflow steps.

    • Recovery: Rewrite description to trigger-only wording.

  • Hidden assumptions: Skill assumes unavailable tools.

    • Recovery: Add tool requirements and fallback behavior.

  • Non-deterministic output: Different agents produce incompatible reports.

    • Recovery: Add a strict output template.

  • Script dependency lock-in: Team cannot verify without local script execution.

    • Recovery: Use AI-guided checklist mode and keep the same report format.

Output Format

Return this report:

Skill: <path>
Discovery: PASS <n>, PARTIAL <n>, FAIL <n>
Content: PASS <n>, PARTIAL <n>, FAIL <n>
Quality: PASS <n>, PARTIAL <n>, FAIL <n>
Critical issues:
- <issue>
Required fixes:
1) <fix>
Decision: READY | NEEDS REVISION

References

  • references/checklist.md
  • docs/WEB-CHAT-PACK.md

What ships with it: 1 file

1.1 KB alongside SKILL.md

references/

Keep looking

Skills are one crate of 326,835. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.