agentsclimarketplace

Audit runtime policy

Skill Mergemat/agent-readiness-skills/skills/audit-runtime-policy

Harness-agnostic skills for agent-readable repos and task prep

Install
npx -y skills add Mergemat/agent-readiness-skills --skill audit-runtime-policy

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

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 0 stars0 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.

What its author says it does

Copied from the file, not written here

Audits tool, connector, skill, hook, side-effect, and approval policy for a repo or task. Use when user asks what tools are allowed, whether MCP/connectors/skills may be used, or wants runtime-policy gaps before agent execution.

SKILL.md

1.3 KB, 241 tokens by cl100k_base, as published. Nobody here has run it

Audit Runtime Policy

Check what an agent may do in this repo or task.

This skill audits policy. It does not grant permissions.

Inspect

Look for:

  • root agent instructions
  • docs/agents/tool-policy.md or equivalent
  • task-specific instructions
  • CI, hooks, pre-commit, lint-staged, or workflow files
  • connector, MCP, and skill configuration
  • approval rules for writes, network, deploy, secrets, and destructive actions

Use policy-format.md for formal policy output.

Classify

For each requested or available capability, classify:

  • allowed
  • allowed_with_approval
  • read_only
  • forbidden
  • unknown

Also classify risk:

  • local read
  • local write
  • network read
  • external write
  • secrets/data
  • destructive
  • deploy/runtime

Output

Return:

  • allowed tools and limits
  • blocked or approval-gated actions
  • missing policy facts
  • required checks or hooks
  • deviations from task or repo policy

If policy is missing, recommend the smallest docs/agents/tool-policy.md entry needed.

What ships with it: 1 file

522 B alongside SKILL.md

references/

Keep looking

Skills are one crate of 326,984. 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.