codex-review-squash-message
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted commit messages and pull request data which could contain instructions intended to bias the agent. Ingestion points: The context gathering script reads the draft squash message and fetches PR data from GitHub. Boundary markers: The script uses explicit section headers to delimit inputs. Capability inventory: The skill utilizes git and gh for read-only context gathering. Sanitization: No specific sanitization is performed on external content.
- [EXTERNAL_DOWNLOADS]: The skill fetches pull request details and diffs from GitHub. Evidence: Execution of gh pr view and gh pr diff in scripts/context.sh.
- [COMMAND_EXECUTION]: The skill executes local git and gh commands to retrieve necessary repository information. Evidence: Use of git rev-parse and the gh CLI within the provided bash script.
Audit Metadata