review-ai-writing
Pass
Audited by Gen Agent Trust Hub on May 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses standard shell commands such as
git diff,find,git log, andmkdir, along with the GitHub CLI (gh pr view), to discover files and retrieve commit/PR metadata. These actions are standard for developer tools and are limited to the local repository environment.- [PROMPT_INJECTION]: The skill has an inherent surface for Indirect Prompt Injection (Category 8) because it analyzes untrusted text from the codebase and PR metadata. - Ingestion points: Processes text from various source files (Markdown, Python, TypeScript, etc.), commit messages, and PR descriptions.
- Boundary markers: Absent. The instructions do not specify the use of delimiters or 'ignore' commands when passing ingested content to subagents for analysis.
- Capability inventory: Spawns subagents via the
Tasktool and writes results to a local JSON report at.beagle/ai-writing-review.json. - Sanitization: No sanitization or escaping is applied to the ingested content before processing.
Audit Metadata