voice_command_to_skill
Maps natural language voice commands to concrete LabClaw skill invocations. Parses ASR output, identifies intent, selects target skill, fills parameters from context, and provides prompt templates — enabling hands-free, voice-driven anywhere-lab experiences where researchers control analysis, guidance, and data export by speaking.
W011: Third-party content exposure detected (indirect prompt injection risk).
What this means
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
Why it was flagged
Runtime path ingests the outsider-authored ASR transcript (“Raw ASR transcript (string)”) from the user’s voice, then uses it directly for intent recognition and constructs prompt templates including a fallback that explicitly incorporates the transcript.
- Repository
- wu-yc/LabClaw
- Commit
df37802
- Audited
- Security analysis
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.