create-pull-request
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection vulnerability surface because it processes untrusted data from git history and repository contents.
- Ingestion points: Data enters the agent's context through commands like
git log origin/main..HEADandgit diff origin/main..HEADinSKILL.md. - Boundary markers: There are no explicit instructions or delimiters telling the agent to ignore prompt injection instructions embedded inside git commit messages or code diffs.
- Capability inventory: The skill possesses repository write and interaction capabilities via
git push origin HEADandgh pr create. - Sanitization: The skill mitigates basic shell-level injection risks by instructing the agent to save the PR body to a temporary file (
/tmp/pr-body.md) and use the--body-fileflag instead of passing raw strings directly to the shell command.
Audit Metadata