wiki

Pass

Audited by Gen Agent Trust Hub on Apr 7, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill behaves as described, providing a local markdown-based wiki for knowledge persistence without any malicious intent or hidden behaviors.\n- [DATA_EXPOSURE]: All file operations are contained within the project-local .omc/wiki/ directory. The skill does not attempt to access credentials, SSH keys, or other sensitive system locations.\n- [INDIRECT_PROMPT_INJECTION]: The skill facilitates reading data that was previously written to the wiki. This creates a standard surface for indirect prompt injection common to all knowledge-retrieval tools.\n
  • Ingestion points: Wiki page content processed via wiki_ingest and wiki_add (SKILL.md).\n
  • Boundary markers: None defined in the interface description.\n
  • Capability inventory: wiki_read, wiki_query, and wiki_list allow the agent to read stored content (SKILL.md).\n
  • Sanitization: No sanitization or validation of the ingested markdown content is described.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 7, 2026, 06:36 AM