wiki-adapter-prd

Pass

Audited by Gen Agent Trust Hub on Jul 27, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXFILTRATION]: The skill incorporates a mandatory sensitive content scanner that identifies and blocks the storage of live credentials, tokens, or private keys, effectively preventing accidental data exposure.
  • [EXTERNAL_DOWNLOADS]: URL ingestion is strictly governed by a fetch-guard mechanism, ensuring that external content is vetted before processing.
  • [PROMPT_INJECTION]: Although the skill processes untrusted external documents (an indirect injection surface), it reduces the risk by using strict structural mapping and a final validation step for all generated pages.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 27, 2026, 11:30 AM
Security Audit — agent-trust-hub — wiki-adapter-prd