exo-distributed
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Recommends installing the
exo-explorepackage viapipto enable distributed inference capabilities.- [COMMAND_EXECUTION]: Utilizes theexocommand-line utility for peer discovery, cluster management, and running inference benchmarks.- [INDIRECT_PROMPT_INJECTION]: The skill sets up an OpenAI-compatible API surface that processes user-supplied prompts. - Ingestion points: API requests to
http://localhost:8080/v1/chat/completionsinSKILL.md. - Boundary markers: Not explicitly defined; uses default LLM prompting behavior.
- Capability inventory: Local LLM inference execution across connected devices.
- Sanitization: Relies on the inference engine's internal safety measures; no external sanitization logic is provided in the skill.
Audit Metadata