hummbl-io/agent-skills/agent-audit
Claude Code and agent skill by hummbl-io/agent-skills
Trust tier 2
- community
Source check pending
Checking that the SKILL.md file is still in the repository.
Audit a specific agent's bus activity, commit history, trust score, and guardrail compliance.
Install
npx vskill@latest install hummbl-io/agent-skills/agent-auditSKILL.md
The same SKILL.md was published earlier as hummbl-io/agent-skills/agent-audit.
Agent Audit Command
Deep-dive audit of a specific agent's behavior, trust, and compliance.
Execution
Given $ARGUMENTS as the agent name, run ALL of these in parallel:
1. Bus Activity
2. Commit History
3. Guardrail Check
- Read the agent's guardrails file if it exists:
~/.agents/rules/gemini-guardrails.md~/.agents/rules/_archived/kimi-guardrails.md- Check approved identities, approved scope, blocked scope
- Verify recent bus messages use approved identity
- Check for prohibited message types (DECISION, DIRECTIVE)
4. Trust Assessment
- Count bus messages (volume)
- Check for BLOCKED messages from this agent
- Check for ERROR messages mentioning this agent
- Look for audit findings or revert commits referencing this agent
- Check AGENTS.md for current status (active/probation/retired)
Output Format
Known Agents
codex-- trusted engineering and critical-review peer; no standing primary authoritydevin-- supervised delegated runtime; ACTIVE-AIP/PROBATIONARY until the
canonical roster says otherwise
claude-code-- on-demand, quota-constrained specialist; never assume it is
the primary/default auditor
agy-- on-demand Google AI Pro GitOps specialist; distinct fromgeminigemini-- separate probationary research identity- System agents as configured in your agent registry
Skill Chains
- For delegation history complements bus and commit data ->
[cross-runtime-bridge](python ~/bin/cross_runtime_bridge.py sessions)
Security
- Security scanPassed, 100 of 100Pattern checks on the skill's files.
- Model reviewNot runA model review is scheduled for the next rescan.
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/agent-audit/SKILL.md
Popularity
- 7-day trend score
- 5
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/hummbl-io/agent-skills/agent-audit)