ClawAudit verdict
omnivoice
The skill uses various dependencies, including Whisper, Speaker ID, and CosyVoice2, which may pose security risks if not properly validated. Additionally, it requires users to store voice references and credentials, which could be sensitive information.
⚠ 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 (3)
Pipe to python — executes piped content as Python code
scripts/feishu_send_audio.sh · prose · downgraded · | python3
Uses eval() — can execute arbitrary code
scripts/voice_identify.py · prose · downgraded · eval(
Python os.environ.get — reads environment variable
scripts/cosyvoice_clone.py · prose · downgraded · os.environ.get(
Permissions & capabilities
No declared permissions — minimal attack surface.
Is this flag fair?
Thanks — recorded.