paniolo-writing-style-best-practices
Pass
Audited by Gen Agent Trust Hub on Jul 21, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill processes external text files (READMEs, wiki pages, code comments), which constitutes an indirect prompt injection surface. The risk is assessed as negligible because the skill's capabilities are restricted to reading files, and it includes explicit instructions to avoid fabricating content.
- Ingestion points: Prose within external files such as wiki authoring pages, README.md, and code comments as defined in SKILL.md.
- Boundary markers: Includes negative constraints like 'Do not fabricate content' and 'Do not rewrite content outside what the user provided or requested'.
- Capability inventory: The skill is scoped to
file-read. There is no evidence of network access, shell execution, or file-system writing capabilities. - Sanitization: No specific filtering or sanitization of ingested content is described, relying on model guardrails and explicit task scoping.
Audit Metadata