Connectors

Plugged into Dust today, open tomorrow

Each space holds one or more connectors, with a default one and per-dataset agent or key overrides.

  • Dust

    Full benchmark and inspection: agents, instructions, tools, skills, usage logs, credits and credit price.

    Available
  • CSV import

    Results from any system (Glean, Copilot, internal RAG…) with time, credits or cost, linked by row code.

    Available
  • ChatGPT

    Benchmark the assistants and GPTs of your organisation.

    Coming soon
  • Claude

    Benchmark Claude projects and agents.

    Coming soon
  • MCP

    Any agent exposed through the Model Context Protocol.

    Coming soon

Custom

Dedicated connectors for on-premise generative AI platforms

Is your generative AI platform hosted in-house? We build custom connectors to benchmark, inspect and govern your agents, wherever they run.

Discuss your platform

Your platform isn't listed? Tell us which one to prioritise. Contact →

Your AI agents deserve serious measurement

Connect your Dust workspace and get a first inspection and a first benchmark within minutes.