uk-witness-statement-review

Pass

Audited by Gen Agent Trust Hub on Jul 8, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill operates as a document review tool with no evidence of malicious intent or unauthorized data access. It focuses on analyzing user-provided legal content for professional accuracy.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it is designed to process untrusted external documents such as draft witness statements, correspondence, and exhibits.
  • Ingestion points: User-supplied draft statements and document uploads (SKILL.md, README.md).
  • Boundary markers: While the skill does not use explicit character delimiters, it instructs the agent to perform a structured classification of the input text (statement mapping), which provides structural context for the AI.
  • Capability inventory: The skill leverages browser and search capabilities to retrieve public legal rules and authorities (SKILL.md).
  • Sanitization: No specific input sanitization is mentioned, which is typical for analysis-focused tools relying on classification logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 8, 2026, 08:53 AM
Security Audit — agent-trust-hub — uk-witness-statement-review