matter-briefing

Pass

Audited by Gen Agent Trust Hub on May 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: Local Data Retrieval: The skill directs the agent to read case-specific markdown and YAML files from a defined local directory structure. This is a standard data processing task for summarizing legal records.
  • [SAFE]: No Dangerous Capabilities: Analysis of the workflow shows no instructions for network access, external downloads, or shell command execution. The skill explicitly states it is read-only and does not perform write operations or strategy recommendations.
  • [SAFE]: Logical Security Controls: The inclusion of a 'Conflict Gate' serves as a functional safety check, preventing the agent from generating summaries for unregistered cases and ensuring compliance with a predefined intake process.
  • [SAFE]: No Injection or Obfuscation: The skill instructions are clear and do not contain patterns related to prompt injection, obfuscated URLs, or hidden malicious payloads.
Audit Metadata
Risk Level
SAFE
Analyzed
May 15, 2026, 07:07 AM
Security Audit — agent-trust-hub — matter-briefing