sync-models

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • Command Execution: The skill utilizes a shell command to execute an embedded Python script that calculates differences between local configuration files and the remote router state. This script also invokes curl to fetch data from https://router.huggingface.co/v1/models, which is a vendor-owned resource associated with the skill's core functionality.
  • Configuration Management: The skill is designed to read and modify environment configuration files (chart/env/prod.yaml and chart/env/dev.yaml) to add new model definitions or prune deprecated ones. These actions are performed to keep the chat interface's model catalog synchronized with the live backend.
  • Remote Resource Interaction: The skill involves fetching model metadata from the HuggingFace router and performing web research on third-party provider documentation (e.g., Groq, Together, DeepInfra) to identify model capabilities like reasoning support and parameter counts. This information is used to populate configuration flags such as supportsReasoning and supportsArtifacts.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 06:06 AM
Security Audit — agent-trust-hub — sync-models