Research agents
Agents that gather from several sources, compare them, and write the considered answer with the evidence it rests on.
Compare these sources and tell me where they disagree, with the passage for each.
A written comparison, the passages it rests on, and a list of what could not be established.
What a research agent is designed to do
Some questions have no single document behind them. A case file, two years of correspondence and three amendments have to be read together before anyone can say what the position is. That reading is a day of work, it is done by a different person each time, and no two write-ups have the same shape.
The pack is designed to:
- read the sources that bear on the question, and say which ones were read;
- show where the sources disagree, and quote both;
- separate what is established from what is inferred, and name what is missing.
Agents answer from your own regulations, documents and records, with the source attached.
The primitives it composes
An agent is a model, a set of approved skills, tools and a policy, applied to a scope.
Agents reach your systems through approved connectors — APIs, databases, applications — each with its own permissions.
Changes wait for human approval before they run.
What it runs on
Skills come from a signed vendor catalogue or your own uploads, and can be enabled or disabled per workspace.
Every answer shows the exact commands that ran, their exit codes and output, per host.
Bring your own model — OpenAI-compatible endpoints, Ollama, Anthropic, or Cloudflare AI Gateway / Workers AI — with keys envelope-encrypted at rest.
Customer-controlled inference and execution boundaries.
Bring one question of your own.
Request a demo