learn-from-code-review
Pass
Audited by Gen Agent Trust Hub on Mar 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes
gh(GitHub CLI) commands to retrieve repository information and list pull requests. This is a standard operation for the skill's intended purpose. - [EXTERNAL_DOWNLOADS]: The skill communicates with GitHub's official API (api.github.com) to download pull request comments and review data. This is a legitimate use of a well-known service.
- [PROMPT_INJECTION]: The skill processes untrusted input from pull request comments, creating a surface for indirect prompt injection.
- Ingestion points: External comments and reviews are retrieved from the GitHub API in
SKILL.md(Step 2). - Boundary markers: There are no explicit delimiters or instructions to treat comment text as untrusted data during the distillation phase.
- Capability inventory: The skill can write files to the local filesystem (creating new skills) and create pull requests using the
create_prtool. - Sanitization: The workflow filters for bots and short strings but does not specifically sanitize the content of comments for malicious instructions.
Audit Metadata