ClawAudit verdict
evoclaw
This skill implements a self-evolving agent identity framework that tracks experiences, proposes SOUL.md updates, and requires human governance approval before changes; all operations are local file reads/writes with explicit human oversight built into the governance pipeline.
⚠ Flagged for review — coarse, uncorroborated signal, not a confirmed exploit. Review the config yourself before installing.
Automated static analysis — not a human review. ClawAudit flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.
Findings (8)
Possible hardcoded credential
references/sources.md · code · API_KEY="${!api_key_env}
Pipe to python — executes piped content as Python code
references/heartbeat-debug.md · code · | python3
Instructs covert action — may act without user awareness
configure.md · code · silently
Accesses OpenClaw config/secrets directly
references/heartbeat-debug.md · code · ~/.openclaw/openclaw.json
<script> tag in markdown — potential code injection
tools/soul-viz.py · prose · downgraded · <script>
References agent memory files
configure.md · code · MEMORY.md
subprocess execution — runs system commands from Python
validators/run_all.py · prose · downgraded · subprocess.run(
Popular HTTP library — network access
validators/validate_experience.py · prose · downgraded · got
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class 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?
Thanks — recorded.