finta-reference-architecture

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a technical design guide for integrating Finta CRM into a fundraising stack, focusing on architecture patterns rather than executable scripts.
  • [SAFE]: Instructions explicitly advocate for security best practices, including using separate credentials for different environments, enforcing least-privilege access, and redacting sensitive data from logs and diagnostics.
  • [SAFE]: The provided TypeScript code snippets demonstrate safe architectural patterns such as utilizing queues for background tasks, implementing caching strategies, and handling API rate limits with exponential backoff.
  • [SAFE]: External resources point to the official Finta website and blog, which are relevant and trusted sources for the skill's context.
  • [SAFE]: YAML frontmatter restricts the agent's tool access to 'Read' and 'Grep', which is appropriate for a documentation-centric skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 03:45 AM
Security Audit — agent-trust-hub — finta-reference-architecture