legal-diagram

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [SAFE]: The skill implements a secure two-pass extraction architecture. Pass 1 is a deterministic Python engine that harvests legal entities without LLM involvement. Pass 2 uses the LLM for filling specific gaps via a validated JSON Patch mechanism.
  • [EXTERNAL_DOWNLOADS]: Fetches the Mermaid JS engine from the well-known jsDelivr CDN (cdn.jsdelivr.net) using urllib.request.urlopen in scripts/fetch_mermaid.py. This is used to enable offline HTML diagram exports and is a standard dependency fetch.
  • [COMMAND_EXECUTION]: Executes the Mermaid CLI (mmdc) via subprocess.run in scripts/verify_render.py for diagram verification. This is a local diagnostic tool and does not process untrusted shell arguments.
  • [SAFE]: The skill processes user documents but includes explicit instructions and architectural guards, such as the patch_gate.py validator, to mitigate risks associated with untrusted data processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 07:33 PM
Security Audit — agent-trust-hub — legal-diagram