sop-build

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.80). These URLs point to third‑party GitHub repositories (personal account) that the skill instructs users to git-clone into a local skills directory—installing unvetted code from unknown repos can be used to distribute malware.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). In tasks/from-recording.md (and similarly tasks/interview.md), the agent consumes user-supplied free-text content such as pasted transcripts or inline SOP intake and then writes/updates SOP content, so outsider-authored text can be ingested without selecting a specific trusted item first.