technical-writing
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists entirely of instructional markdown and reference files providing stylistic and factual guidelines for technical writing. It contains no code, scripts, or executables.
- [SAFE]: No network operations, credential usage, or sensitive file access patterns were detected. The skill instructions specifically advise against including sensitive internal metadata like ticket keys in user-facing content.
- [PROMPT_INJECTION]: The skill includes a 'Rule precedence' section that explicitly prioritizes safety rules above author style choices, ensuring the agent maintains its safety guardrails during the writing process.
- [INDIRECT_PROMPT_INJECTION]: The skill acts on external documentation, forming an ingestion surface. Mandatory Evidence Chain: 1. Ingestion points: technical documents, specs, READMEs; 2. Boundary markers: None specified; 3. Capability inventory: The skill itself defines no tool access or network capabilities; 4. Sanitization: Grounding claims and review passes.
Audit Metadata