repo-review
Pass
Audited by Gen Agent Trust Hub on Jul 16, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: Indirect Prompt Injection Surface. Ingestion points: The skill analyzes untrusted repository content, including source code, commit history, and specific agent-facing files like AGENTS.md (SKILL.md, references/worktree-checklist.md). Boundary markers: The guidelines mention using repository-specific guidance but do not establish formal markers or delimiters to isolate untrusted content. Capability inventory: The skill is authorized to perform repository checks and execute validation commands (references/checklist.md). Sanitization: The instructions do not specify a requirement for sanitizing or escaping the content retrieved from the repository before processing.
- [COMMAND_EXECUTION]: Risk from Repository-Defined Validation Tools. The skill instructs the agent to run non-mutating checks such as type checkers, linters, and build system commands (references/worktree-checklist.md). Executing these tools on a dirty worktree or untrusted codebase carries an inherent risk of arbitrary code execution if the repository's configuration (e.g., justfile, package.json) is maliciously crafted.
Audit Metadata