agently-triggerflow-model-integration

Pass

Audited by Gen Agent Trust Hub on Mar 11, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists exclusively of technical documentation in markdown format. No executable scripts, binaries, or configuration files are included in the package.
  • [EXTERNAL_DOWNLOADS]: The documentation references official Agently repositories and documentation sites (github.com/AgentEra/Agently and agently.tech). These are legitimate vendor resources and well-known documentation services used for developer guidance.
  • [PROMPT_INJECTION]: The skill describes patterns for processing LLM outputs through streaming. Ingestion points: references/streaming-and-dispatch.md (delta and instant streaming). Boundary markers: Not explicitly specified in the architectural guide. Capability inventory: references/multi-request-patterns.md (dispatching events, routing with when(), and triggering further model requests). Sanitization: Not explicitly discussed. The skill provides best practices to mitigate surface risks by recommending bounded concurrency and avoiding unbounded task spawning.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 11, 2026, 05:52 PM