code-review
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from git diffs and external spec files, which could contain malicious instructions designed to manipulate the review output. * Ingestion points: Git diff hunks, commit logs, and documentation files matching docs/, specs/, or .scratch/ (SKILL.md). * Boundary markers: Absent; no specific delimiters or ignore-instruction warnings are defined to separate code content from agent instructions. * Capability inventory: Local git execution and spawning of parallel sub-agents via the review_run tool (SKILL.md). * Sanitization: Absent; the skill directly interpolates raw file and diff content into sub-agent prompts without escaping or validation.
- [COMMAND_EXECUTION]: The skill executes standard git commands to perform its task. * Evidence: Usage of git diff, git log, and git rev-parse to analyze repository state (SKILL.md). These are legitimate operations for a code review tool.
Audit Metadata