agent-init
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- SAFE (SAFE): The skill is a documentation and templating utility that does not contain malicious code, remote downloads, or obfuscation. It is designed to improve the operational clarity and safety of other agents.
- COMMAND_EXECUTION (SAFE): The skill suggests running basic filesystem commands like 'ln -s' to create symlinks for agent compatibility (e.g., mapping CLAUDE.md to AGENTS.md). This is a standard and safe operation.
- PROMPT_INJECTION (SAFE): The skill facilitates the creation of instruction files that influence downstream agent behavior. 1. Ingestion points: User-provided repository facts during the brainstorming workflow. 2. Boundary markers: The generated templates explicitly include 'Always / Ask first / Never' sections to define agent limits. 3. Capability inventory: File system writes and symlink creation. 4. Sanitization: The skill does not perform automated sanitization of brainstorming input, relying on user review of the resulting markdown documentation.
Audit Metadata