pr-size-breakdown
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted metadata from GitHub Pull Requests, including file paths and change statistics, creating a surface for potential indirect prompt injection. 1. Ingestion points: PR metadata and file lists fetched via the gh CLI as described in SKILL.md. 2. Boundary markers: No specific delimiters are defined to separate external data from internal logic. 3. Capability inventory: The skill performs analytical tasks and generates reports without dangerous capabilities like file writing or code execution. 4. Sanitization: No sanitization of external metadata is mentioned.
- [COMMAND_EXECUTION]: The skill instructs the use of the GitHub CLI (gh) for fetching Pull Request data via the GraphQL API, which is standard for repository management tasks.
- [EXTERNAL_DOWNLOADS]: The skill retrieves metadata from GitHub's official API, a well-known and trusted service.
Audit Metadata