doc-coauthoring
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security issues were detected. The skill outlines a multi-stage workflow for collaborative writing and document refinement.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided context to generate documentation (Stage 1 and 2), which is an inherent part of its functionality. While Stage 3 instructs the agent to "pretend you have no access to prior conversation" to simulate a fresh reader, this is a localized instruction for quality assurance and does not target safety filters or sensitive data. The skill has no associated tools or scripts that could be leveraged for data exfiltration or command execution.
Audit Metadata