triage
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from GitHub issue bodies and comments, creating a potential surface for indirect prompt injection attacks.
- Ingestion points: The skill reads the body and all comments of a specified GitHub issue to assess implementation readiness (SKILL.md).
- Boundary markers: The instructions include a clear boundary stating that issue bodies and comments must be treated as untrusted context and should not override repository guidance or authorize actions (SKILL.md).
- Capability inventory: The skill is capable of modifying GitHub issues using the
gh issue editcommand (SKILL.md). - Sanitization: No automated sanitization is mentioned; the skill relies on the AI model's compliance with the provided instructions to mitigate injection risks.
- [COMMAND_EXECUTION]: The skill invokes the
ghcommand-line tool to perform authenticated updates to GitHub issues. These operations are restricted to authorized fields and require user approval unless previous authorization was granted (SKILL.md).
Audit Metadata