ClawAudit verdict
yuyonghao-multimodal-base
Multimodal pipeline using OpenAI GPT-4V and Whisper APIs with user-supplied key for image/audio analysis; capabilities are used transparently for the stated purpose.
⚠ 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.
Permission integrity
package_install
Findings (8)
Possible hardcoded credential
SKILL.md · code · ApiKey: 'your-api-key
Accesses process.env — reads environment variables
SKILL.md · code
Accesses sensitive environment variables
SKILL.md · code · process.env.OPENAI_API_KEY
Dynamic import() — loads module at runtime
src/image-processor.js · prose · downgraded · import('
References child_process — can spawn system processes
src/speech-recognizer.js · prose · downgraded · child_process
Uses spawn() — can execute external programs
src/speech-recognizer.js · prose · downgraded · spawn(
Popular HTTP library — network access
package-lock.json · prose · downgraded · axios
POSTs data to external URL
src/speech-recognizer.js · prose · downgraded · .post(
'https://
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_installcredential_access Is this flag fair?
Thanks — recorded.