modellix

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONCREDENTIALS_UNSAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the modellix-cli tool to execute various operations, including model runs, task status checks, and data downloads.
  • [EXTERNAL_DOWNLOADS]: The skill fetches task results, request schemas, and documentation from external endpoints at api.modellix.ai, docs.modellix.ai, and the Modellix/modellix-plugin GitHub repository.
  • [DATA_EXFILTRATION]: User-provided prompts and source media are transmitted to the Modellix API (api.modellix.ai) and CDN for processing, which is the primary intended function of the skill.
  • [CREDENTIALS_UNSAFE]: The skill manages a MODELLIX_API_KEY for authentication. It includes specific security notes advising the agent never to print secrets in logs and to prefer session-scoped key usage to minimize exposure.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 07:28 PM
Security Audit — agent-trust-hub — modellix