OpenClaw代码审查助手
Analysis
The skill advertises real code review and security scanning, but the included script mostly prints canned example findings unrelated to the user's code.
Findings (3)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.
echo "📁 Reviewing: $TARGET" ... cat << 'REPORT' ... "File: example.js" ... "✅ Review complete!"
The review branch accepts a user target but then emits a hard-coded report about example.js; no artifact code reads or analyzes the target, contradicting the advertised review/security-scanning purpose.
"ownerId": "kn754w8negkve53dntzhycyd0h82565k", "slug": "code-review-assistant", "version": "1.0.0"
The bundled metadata differs from the supplied registry listing, which names a different owner ID, slug, and version; this creates package provenance/version ambiguity.
Checks whether tool use, credentials, dependencies, identity, account access, or inter-agent boundaries are broader than the stated purpose.
echo "⚠️ GitHub integration not configured"; echo " Set GITHUB_TOKEN to enable PR reviews"
The script asks for a GitHub token for PR reviews, while the provided metadata declares no required environment variables or primary credential; the code shown does not actually use the token.
