technical-writing
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No security issues or malicious patterns were detected. The skill exclusively provides architectural and stylistic guidelines for writing technical documentation, such as Diátaxis and the Google Developer Documentation Style Guide.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to review and rewrite untrusted external data such as RFCs, readmes, and PR descriptions. While this establishes a vulnerability surface for indirect instructions, the skill does not grant the agent any privileged tools or executable capabilities that could be exploited. 1. Ingestion points: Documents, RFCs, readmes, and PR descriptions processed in SKILL.md. 2. Boundary markers: None specified. 3. Capability inventory: No scripts, subprocess calls, file-system writes, or network operations are present. 4. Sanitization: None.
Audit Metadata