ascendc-operator-doc-writer
Pass
Audited by Gen Agent Trust Hub on Aug 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill performs standard technical documentation tasks by reading and summarizing local AscendC source files. The workflow is well-defined and aligns with its stated purpose of assisting engineers.- [PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection by processing untrusted source code and reference templates provided by the user. While it lacks robust sanitization or delimiters, it instructs the model to ground its output in confirmed facts from the source code. Ingestion points include the operator's host, tiling, and kernel implementation files. The capability inventory is restricted to local file reading and searching using standard tools.
Audit Metadata