Caveat verdict
counterclaw
counterclaw-core
Defensive security middleware that scans inputs/outputs for prompt injection and PII; logs violations locally only with optional email integration clearly documented.
⚠ 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.
Permission integrity
package_install
Findings (7)
Prompt injection — tries to override agent instructions
SKILL.md · code · ignore previous instructions
Pipe to python — executes piped content as Python code
README.md · code · | python3
Accesses shell history/config
README.md · code · ~/.zshrc
Accesses OpenClaw config/secrets directly
README.md · code · ~/.openclaw/.env
References agent memory files
SKILL.md · frontmatter · MEMORY.md
Possible prompt injection — attempts to redefine agent identity
tests/test_email_protection.py · prose · downgraded · You are now
Python os.getenv — reads environment variable
src/counterclaw/middleware.py · prose · downgraded · os.getenv(
Permissions & capabilities
No declared permissions — minimal attack surface.
package_install Is this flag fair?
Thanks — recorded.