Documents

skill-trust-auditor

Audit a named ClawHub skill or skill URL before installation by combining OpenClaw verification with bounded static analysis. Use when the user explicitly asks whether a skill is safe or requests a pre-install review; report evidence and uncertainty instead of treating a score as proof.

What it does

Audit a named ClawHub skill or skill URL before installation by combining OpenClaw verification with bounded static analysis. Use when the user explicitly asks whether a skill is safe or requests a pre-install review; report evidence and uncertainty instead of treating a score as proof.

The skill document

Skill Trust Auditor

Audit any ClawHub skill for security risks before installation.

🛠️ Installation

Tell OpenClaw: "Install @jonathanjing/skill-trust-auditor."

2. Manual Installation (CLI)

If you prefer the terminal, run:

openclaw skills install @jonathanjing/skill-trust-auditor

Setup (first run only)

bash "{baseDir}/scripts/setup.sh"

Audit a Skill

First request the registry trust envelope:

openclaw skills verify @owner/skill --json

Then run bounded local analysis when the user asks for a deeper audit:

bash "{baseDir}/scripts/audit.sh" [skill-name-or-url]
# Example:
bash "{baseDir}/scripts/audit.sh" steipete/clawhub
bash "{baseDir}/scripts/audit.sh" https://clawhub.ai/someuser/someskill

Output:

{
  "skill": "someuser/someskill",
  "trust_score": 72,
  "verdict": "INSTALL WITH CAUTION",
  "risks": [
    {"level": "HIGH", "pattern": "curl to external domain", "location": "scripts/sync.sh:14"},
    {"level": "MEDIUM", "pattern": "reads MEMORY.md", "location": "SKILL.md:23"}
  ],
  "safe_patterns": ["no env var access", "no self-modification"],
  "author_verified": false,
  "recommendation": "Review scripts/sync.sh:14 before installing. The external curl call could exfiltrate data."
}

Post to user with clear summary:

🛡️ Trust Audit: someuser/someskill
Score: 72/100 — ⚠️ INSTALL WITH CAUTION

🔴 HIGH: curl to unknown domain in scripts/sync.sh:14
🟡 MEDIUM: reads your MEMORY.md

Recommendation: Inspect line 14 of sync.sh before proceeding.
Inspect the exact version's Files tab or installed folder before proceeding.

Trust Score Guide

ScoreVerdictAction
90-100✅ SAFEInstall freely
70-89⚠️ CAUTIONReview flagged items first
50-69🟠 RISKYOnly if you understand the risks
0-49🔴 DO NOT INSTALLHigh probability of malicious intent

Risk Pattern Reference

HIGH RISK (-30 each):

  • process.env access in scripts
  • curl/wget to non-standard domains
  • Reading ~/.config or ~/.openclaw directly
  • exec() with user-controlled input
  • Instructions to modify SOUL.md/AGENTS.md/openclaw.json

MEDIUM RISK (-10 each):

  • Any outbound API calls (even to known services)
  • File writes outside workspace
  • Reading MEMORY.md or diary files

LOW RISK (-3 each):

  • web_fetch to standard domains
  • Read-only file access in workspace

Do not auto-install from an audit score. Present findings, the registry decision, and the exact install reference; let the user authorize installation separately.

ClawHavoc Pattern Reference

See {baseDir}/references/clawhavoc-patterns.md for known malicious patterns. Treat it as a heuristic reference, not a complete malware signature set.

Related skills

ClawHub reputation checker for clawsec-suite. Adds a standalone reputation gate before guarded skill installation.

34 installs

Review your installed agent skills to see what you actually use, what overlaps, and what may no longer be worth keeping.

49 installs69 stars

Agent skill quality checker. Input a skill directory or skill files; output trigger clarity, metadata issues, examples, safety boundaries, installability, po...

36 installs

Get a structured risk report on a skill package before installing or publishing it.

128 installs

Audit an OpenClaw agent workspace and generate standardized evaluation reports, scores, and patches. Use when asked to review memory quality, retrieval effic...

32 installs

Use when checking for security vulnerabilities in NanoClaw skills, before installing new skills, or when asked about security advisories affecting the bot

48 installs