create-pr
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data which could contain malicious instructions.
- Ingestion points: Analyzes commit messages and diffs in SKILL.md via
git logandgit diffto generate pull request content. - Boundary markers: Absent. The instructions do not include delimiters or specific warnings to ignore instructions that might be embedded in the analyzed repository data.
- Capability inventory: The skill utilizes
git pushandgh pr createas defined in SKILL.md and its allowed-tools configuration. - Sanitization: Absent. No filtering or sanitization is applied to the analyzed repository data before it is used to generate the PR title and description.
- [COMMAND_EXECUTION]: The skill executes Git and GitHub CLI commands to manage code branches and pull requests. These actions are limited to the specific tools defined in the
allowed-toolsconfiguration, representing a least-privilege approach.
Audit Metadata