docs-impact-localizer
Pass
Audited by Gen Agent Trust Hub on May 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [Command Execution Surface]: The skill uses the
gh(GitHub CLI) tool to performgh pr diffoperations. This is a standard and expected functionality for a tool intended to automate documentation updates based on code changes.\n- [Indirect Prompt Injection Surface]: The skill processes untrusted content from documentation pages and pull request diffs to generate instructions for downstream verification tools.\n - Ingestion points: Ingests content from documentation files (paths in
scope_pages[]) and the PR diff (SKILL.md).\n - Boundary markers: The instructions do not implement specific boundary markers or "ignore instructions" directives for the data being analyzed.\n
- Capability inventory: The skill utilizes shell commands (
gh pr diff) and generates verification commands (e.g.,grep,apm install --help) that are passed to other agent components for execution (SKILL.md).\n - Sanitization: No specific sanitization procedures are described for the data ingested before it is incorporated into the output task brief.
Audit Metadata