together-batch-inference
Pass
Audited by Gen Agent Trust Hub on Mar 30, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a legitimate integration with the Together AI Batch API for high-volume offline inference.
- [EXTERNAL_DOWNLOADS]: Recommends installing official, well-known SDKs (
togetherfor Python andtogether-aifor Node.js) to interact with the service. - [CREDENTIALS_UNSAFE]: Properly instructs users to manage API keys via environment variables rather than hardcoding them within scripts or configuration files.
- [DATA_EXFILTRATION]: Network operations are limited to the official Together AI API endpoints (
https://api.together.xyz) for the purpose of uploading inference data and downloading results. - [COMMAND_EXECUTION]: Scripts perform standard file I/O and network requests without any evidence of arbitrary command execution or dynamic code evaluation from untrusted sources.
Audit Metadata