resonate-human-in-the-loop-pattern-typescript

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [SAFE]: No security issues detected. The skill provides legitimate documentation and code examples for using the Resonate SDK.
  • [EXTERNAL_DOWNLOADS]: The skill references the @resonatehq/sdk and express Node.js packages. These are standard, legitimate dependencies for the documented functionality and originate from the skill's vendor or well-known registries.
  • [PROMPT_INJECTION]: The skill describes patterns for ingesting external human input (e.g., through Express route handlers) to resolve durable promises that control workflow logic. \n
  • Ingestion points: Express route handlers in SKILL.md (e.g., /approve/:promiseId). \n
  • Boundary markers: Data is handled via durable promises with typed responses. \n
  • Capability inventory: Resolved data influences workflow logic, triggering actions through ctx.run() and ctx.rpc(). \n
  • Sanitization: Examples demonstrate sanitization by checking inputs against fixed values (e.g., action === 'accept').
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 06:43 AM
Security Audit — agent-trust-hub — resonate-human-in-the-loop-pattern-typescript