watch-asc-pr
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and act upon data from external sources (pull request reviews and comments) which are potentially attacker-controlled.
- Ingestion points:
SKILL.mdspecifies fetching top-level comments, check results, and GraphQL review threads from a pull request. - Boundary markers: The instructions do not define explicit delimiters or instructions to ignore potential commands embedded within the retrieved pull request content.
- Capability inventory: The skill possesses repository write capabilities, including the ability to implement fixes, commit and push changes, reply to review threads, and potentially trigger merges as described in the workflow.
- Sanitization: There is no mention of sanitization, filtering, or validation of the fetched comments to prevent the agent from being influenced by instructions hidden in the feedback.
Audit Metadata