resonate-recursive-fan-out-pattern-typescript

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill functions as a technical guide for architectural patterns. The provided TypeScript examples demonstrate standard usage of the Resonate SDK for managing distributed workflows and task execution.
  • [INDIRECT_PROMPT_INJECTION]: The "Recursive Research Agent" pattern demonstrates a vulnerability surface where a user-provided topic is interpolated into an LLM prompt. This is a design characteristic of LLM-based agents rather than a malicious implementation.
  • Ingestion points: The topic string parameter in the research generator function (SKILL.md).
  • Boundary markers: None are present in the illustrative SYSTEM_PROMPT provided in the example code.
  • Capability inventory: The example utilizes Resonate SDK primitives (ctx.run, ctx.beginRpc, ctx.detached) to execute workflows and calls the OpenAI API.
  • Sanitization: None is implemented in the code snippet, as it is an architectural example.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 06:42 AM
Security Audit — agent-trust-hub — resonate-recursive-fan-out-pattern-typescript