guided-doc-writing
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes and rewrites user-provided documents, which is an inherent surface for indirect prompt injection. However, the skill lacks the system capabilities (network or file access) required for an attacker to exploit this surface.- [COMMAND_EXECUTION]: The skill's tool rules in AGENTS.md instruct the agent to run the compose-agentsmd utility for rule synchronization. This is a maintenance function using a tool from the same vendor and does not pose a security risk.- [EXTERNAL_DOWNLOADS]: The skill references external rule documentation from the vendor's GitHub repository (metyatech/agent-rules). This is documented neutrally as a configuration source.
Audit Metadata