pr-review-analysis
Pass
Audited by Gen Agent Trust Hub on Apr 29, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from pull request comments. \n
- Ingestion points: Pull request comments are retrieved from external GitHub repositories using
gh apias specified in the Step 1 workflow in SKILL.md.\n - Boundary markers: The instructions do not define boundary markers or explicit isolation for the untrusted comment content within the prompt context.\n
- Capability inventory: The skill uses local file reading and grepping capabilities and can execute
gh apicommands to post replies to the remote platform.\n - Sanitization: No explicit sanitization or filtering of the comment content is described, although the inclusion of a user confirmation step before posting replies acts as a manual checkpoint.
Audit Metadata