llm-wiki-domain-pack
Pass
Audited by Gen Agent Trust Hub on Aug 11, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill is comprised of markdown instructions and a JSON schema; no executable scripts or binaries are present.
- [SAFE]: Detailed analysis across all threat categories found no evidence of malicious intent. Regarding the data processing surface (Category 8):
- Ingestion points: User-provided domain names, purposes, and corpus types defined in the 'Inputs' section of SKILL.md.
- Boundary markers: None explicitly defined in the natural language instructions for input interpolation.
- Capability inventory: Generation of static markdown files (pack.md) and JSON configuration files (schema.overlay.json, profile.json).
- Sanitization: Structural validation and regex patterns for field names are included in the provided JSON schema, and the skill specifies that generated profiles are for review rather than automatic execution.
Audit Metadata