tool-permission-audit
Skill0 starsChanged 5 months agoMITView on GitHub
Use when auditing AI-agent tool permissions, command policies, connector access, GitHub actions, file-system scopes, or automation rules for risk, approval gates, and least-privilege controls.
In this repository19 other files it gives its agents.
- agent-handoff-briefSkill
- api-doc-to-tool-specSkill
- architecture-rfc-reviewSkill
- bug-report-to-reproSkill
- changelog-to-upgrade-planSkill
- competitive-research-briefSkill
- customer-call-to-prdSkill
- database-migration-risk-reviewSkill
- dependency-upgrade-risk-reviewSkill
- incident-log-to-postmortemSkill
- messy-notes-to-decision-logSkill
- privacy-data-flow-mapSkill
- prompt-to-eval-rubricSkill
- repo-launch-readiness-reviewSkill
- roadmap-to-release-planSkill
- security-review-checklistSkill
- support-ticket-to-evalSkill
- telemetry-to-product-insightsSkill
- workflow-automation-scoutSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 14
Use when an agent can read/write files, call connectors, run commands, publish, or touch credentials.
- Distinctive: line 15
Do not use when the agent has no tool access and only produces plain text.
- Distinctive: line 27
Check whether approval gates match the risk.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Tool Permission Audit
Purpose
Review an agent's tools and permission policy for unsafe access, unclear gates, and least-privilege gaps.
Fit
- Use when an agent can read/write files, call connectors, run commands, publish, or touch credentials.
- Do not use when the agent has no tool access and only produces plain text.
Inputs
- Tool list, command allowlist, connector scopes, system instructions, or policy text.
- User expectations for autonomy, external writes, destructive actions, and credentials.
- Known environment restrictions.
Workflow
- Inventory tools and classify their side effects.
- Separate read-only, local write, external write, credential, and destructive actions.
- Check whether approval gates match the risk.
- Identify overbroad scopes and ambiguous wording.
- Recommend a least-privilege policy patch.
Output
Produce Markdown with:
- Permission Matrix
- Risk Summary
- Approval Gates
- Overbroad Access
- Policy Gaps
- Recommended Patch
- Residual Risks
Validation
- Public or external writes are clearly gated.
- Destructive commands are not treated as normal execution.
- Credential handling is explicit.
- Local reversible actions are separated from irreversible actions.
- Recommendations preserve useful autonomy where risk is low.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.