write-agents-md
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is entirely instructional and does not contain executable code, scripts, or commands. It defines a logical framework for human-agent collaboration and documentation structure.
- [PROMPT_INJECTION]: The content focuses on project governance and architecture. There are no instructions intended to bypass safety filters, escalate privileges, or reveal system prompts.
- [DATA_EXFILTRATION]: No network requests, sensitive file access (e.g., .ssh, .env, .aws), or credential markers were found. The skill operates purely within the context of text manipulation and documentation guidance.
- [OBFUSCATION]: The text is plain markdown. There are no Base64 encoded strings, hidden Unicode characters, or homoglyph attacks.
- [REMOTE_CODE_EXECUTION]: The skill does not perform any remote downloads or execute external scripts. It references an internal skill 'clear-writing' for stylistic guidance, which is standard behavior.
Audit Metadata