file-pr
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the GitHub CLI tool (
gh) to perform its primary functions, including creating pull requests and interacting with the GitHub REST API. - [INDIRECT_PROMPT_INJECTION]: The skill processes user prompts and repository data (diffs, Git history, merged PRs) to generate pull request metadata, which presents a surface for indirect instructions to influence agent behavior.
- Ingestion points: User original prompts, repository diffs, recently merged PRs, and Git history (SKILL.md).
- Boundary markers: Absent.
- Capability inventory: Shell execution of
ghCLI commands to create pull requests and call the GitHub API (SKILL.md). - Sanitization: Present; the skill instructions mandate a manual check for secrets and private details in the PR content before submission.
- [SAFE]: The skill follows security best practices by explicitly instructing the agent to audit generated content for credentials or sensitive information before submitting it to a remote repository.
Audit Metadata