file-pr
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data from the repository, specifically git diff output, GLOSSARY.md, and .github/pull_request_template.md, to generate pull request descriptions. This represents an attack surface where a malicious contributor could attempt to influence the agent's behavior through branch content or template files.
- Ingestion points: The skill instructs the agent to read git diff origin/...HEAD and repository-specific configuration files (GLOSSARY.md and PR templates).
- Boundary markers: No explicit delimiters or warnings to ignore embedded instructions are provided for the ingested content.
- Capability inventory: The skill uses git fetch and git diff for reading, and has the capability to create/open pull requests (write access).
- Sanitization: There is no evidence of sanitization or filtering applied to the repository content before it is interpolated into the PR description prompt.
Audit Metadata