ClawAudit verdict

asr-hotwords

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

Mines conversation logs from OpenClaw sessions using local jieba tokenization and an LLM call to build a hotword list, then injects the result back into the agent's voice context; all processing is local/intra-agent with no external exfiltration of conversation content.

⚠ 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.

73
security
50
transparency
70
maintenance

Permission integrity

Installs packages at runtime — transitive dependencies are not auditable

package_install

Findings (3)

Pattern match medium

pip3 install — installs Python packages at runtime

SKILL.md · code · pip3 install

Pattern match medium

Accesses OpenClaw config/secrets directly

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

Pattern match low

Python os.environ.get — reads environment variable

build_words.py · prose · downgraded · os.environ.get(

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.

package_install

Is this flag fair?

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