decision-first-technical-writing
Pass
Audited by Gen Agent Trust Hub on Sep 1, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill contains no executable scripts (Python, Node.js, Shell, etc.). It consists entirely of Markdown instructions for the AI agent and configuration metadata.
- [SAFE]: The instructions focus solely on technical writing best practices, document routing, and consistency checks for architecture documents. No findings were detected in any of the 11 threat categories.
- [PROMPT_INJECTION]: The skill uses phrases like "Important" and "Route to..." but these are used in the context of standard instructional logic for document processing, not to bypass safety filters or override core agent behavior.
- [DATA_EXPOSURE]: There are no patterns of sensitive file access or network operations. The skill explicitly instructs the AI to preserve identifiers like endpoints and error codes from the user's provided source material, which is consistent with its role as a technical writing tool.
- [METADATA_POISONING]: Metadata fields (name, description, author) accurately reflect the skill's purpose without deceptive instructions or hidden payloads.
Audit Metadata