pr
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes shell commands using the GitHub CLI (
gh). It runsgh pr listto fetch recent pull request titles andgh pr createto open new draft pull requests with AI-generated titles and bodies. - [INDIRECT_PROMPT_INJECTION]: The skill represents an attack surface for indirect prompt injection because it processes untrusted data (repository code, diffs, and task inputs) to generate content for pull requests.
- Ingestion points: The skill ingests repository content and agent-generated changes to populate the 'Summary' and 'Evidence' sections of the pull request template.
- Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore potentially malicious instructions embedded in the code being summarized.
- Capability inventory: The skill utilizes the
gh pr createcommand, which performs network operations and writes data to a remote repository. - Sanitization: There is no evidence of sanitization or escaping of the generated content before it is interpolated into the shell command for pull request creation.
Audit Metadata