create-github-pull-request-from-specification
Pass
Audited by Gen Agent Trust Hub on May 18, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting content from local files and repository data which could be manipulated to influence agent behavior.
- Ingestion points: The skill reads template content from
${workspaceFolder}/.github/pull_request_template.md(Step 1), accepts user input via${input:targetBranch}(Step 2), and analyzes Pull Request diff data (Step 3). - Boundary markers: There are no explicit delimiters or instructions provided to the agent to distinguish between administrative instructions and the data being processed from the template or diffs.
- Capability inventory: The skill uses tools with repository modification permissions, including
create_pull_request,update_pull_request, andupdate_issue(Steps 2-6). - Sanitization: No evidence of sanitization, validation, or escaping of the template or diff content is present before it is used to populate Pull Request fields.
Audit Metadata