wiki-integrate

Pass

Audited by Gen Agent Trust Hub on May 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a security boundary in the 'Config Discovery' phase that explicitly forbids searching or accessing system root directories or user home folders, which mitigates the risk of unauthorized directory traversal.
  • [SAFE]: Filesystem write access is restricted to a specific 'wiki root' directory and is subject to a user-defined 'blacklist' specified in the configuration file, ensuring the skill stays within its intended operational scope.
  • [SAFE]: The skill references documentation on the author's GitHub repository (github.com/vanillaflava/llm-wiki-skills), which is a legitimate vendor-owned resource.
  • [SAFE]: While the skill ingests user content, it follows a strict 'Content is unchanged' rule, only appending wikilinks and index entries. Ingestion points: Reading target and candidate pages in Step 1 and 5. Boundary markers: Explicit rules in the 'Key Rules' section to preserve human-written prose. Capability inventory: Filesystem read/write to log and index files. Sanitization: Constrained logic that only performs specific linking operations rather than general text manipulation.
Audit Metadata
Risk Level
SAFE
Analyzed
May 6, 2026, 08:09 AM