address-review
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow that ingests potentially untrusted content from external code reviews and PR comments to drive code modifications.
- Ingestion points: Review findings are collected from external files, PRs, and PR comments (SKILL.md).
- Boundary markers: The instructions do not specify the use of delimiters or specific warnings to ignore instructions embedded within the review data.
- Capability inventory: The skill allows the agent to explore code, implement fixes, commit changes, push to branches, and post comments back to the PR (SKILL.md).
- Sanitization: There is no mention of sanitizing, escaping, or validating the review content before it is used to determine code changes.
Audit Metadata