humanlayer-effect-machine
Pass
Audited by Gen Agent Trust Hub on Oct 19, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a technical documentation file providing guidance on building state machines with the @humanlayer/effect-machine library within the Effect ecosystem. No malicious code, behavioral overrides, or unauthorized activities were detected.
- [EXTERNAL_DOWNLOADS]: The skill references the @humanlayer/effect-machine vendor-specific package and the well-known effect library. These are legitimate, standard dependencies for the library's stated purpose.
- [INDIRECT_PROMPT_INJECTION]: The skill architecture handles external data via state machine events. 1. Ingestion points: Event payloads (SKILL.md). 2. Boundary markers: Integrated via schema definitions. 3. Capability inventory: Effect-based tasks and services defined in the machine (SKILL.md). 4. Sanitization: The library uses effect/Schema for strict runtime parsing and validation. This schema-first approach ensures that any external input is validated against expected types, mitigating the risk of processing malicious instructions embedded in data.
Audit Metadata