Caveat verdict

agent-selector

45
🟠 Risky
Significant risk patterns flagged — automated deep scan, not behavioral proof.

This skill switches agent personalities based on input, and while it has security enhancements, the execution sink detection is a concern.

⚠ Flagged for review — coarse, uncorroborated signal, not a confirmed exploit. Review the config yourself before installing.

Automated static analysis — not a human review. Caveat flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.

0
security
70
transparency
70
maintenance

Findings (10)

Pattern match critical

Accesses sensitive system files

SKILL.md · code · /etc/passwd

Pattern match critical

Possible hardcoded credential

agency-agents/engineering/engineering-technical-writer.md · code · apiKey: 'YOUR_SEARCH_API_KEY

Pattern match high

Uses eval() — can execute arbitrary code

agency-agents/engineering/engineering-ai-engineer.md · prose · downgraded · eval(

Pattern match medium

Accesses sensitive environment variables

agency-agents/engineering/engineering-database-optimizer.md · code · process.env.SUPABASE_ANON_KEY

Pattern match medium

Opens WebSocket connection

agency-agents/specialized/lsp-index-engineer.md · code · WebSocket

Pattern match medium

Base64 encoding/decoding

agency-agents/support/support-infrastructure-maintainer.md · code · base64encode

Pattern match medium

References webhook/callback URL

agency-agents/support/support-infrastructure-maintainer.md · code · WEBHOOK_URL

Pattern match low

Makes HTTP request to external URL

agency-agents/engineering/engineering-feishu-integration-developer.md · code · fetch( 'https://

Pattern match low

References agent memory files

agency-agents/specialized/zk-steward.md · prose · downgraded · MEMORY.md

Pattern match low

Popular HTTP library — network access

agent_selector.py · prose · downgraded · Got

Why the tier is capped

Execution sink present in raw bytes (Hard Floor: class B/D). Final tier capped at Caution — cannot be lifted by any downgrade, example-payload opt-in, or allowlist.

Permissions & capabilities

No declared permissions — minimal attack surface.

Is this flag fair?

Check another skill Browse the registry Auditing your own skills or configs? Use the API