ClawAudit verdict
complex-task-subagent
Reads local files AND makes external network calls
Task orchestration framework for OpenClaw using documented cron and sub-agent APIs; the git clone installs from a named public repo and all functionality matches the stated purpose of multi-stage task automation.
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.
What it does
These are capability combinations: each listed behavior occurs in the skill, but ClawAudit detects co-occurrence — it does not verify that one flows into another. Read the code to confirm a live chain.
Reads local files AND makes external network calls — the capabilities for data exfiltration co-occur (data-flow not verified)
LLM02 · LLM06 · ASI03
Accesses credentials AND makes external network calls — potential credential theft
LLM02 · ASI03
Permission integrity
network_out
file_read
credential_access
Findings (6)
Uses eval() — can execute arbitrary code
references/advanced-patterns.md · code · eval(
<script> tag in markdown — potential code injection
references/advanced-patterns.md · code · <script>
Accesses OpenClaw config/secrets directly
SKILL.md · code · ~/.openclaw/openclaw.json
References SSH/GPG private keys
references/troubleshooting.md · prose · downgraded · ssh-key
Accesses .ssh directory
references/troubleshooting.md · prose · downgraded · .ssh/
Sets world-executable permissions
references/troubleshooting.md · prose · downgraded · chmod 755
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class B). 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.
file_readcredential_accessnetwork_out Thanks — recorded.