ConTree for SWE Agents
======================

Researching SWE agents is hard: there are thousands of heavy-weight environments, setup is slow, and
reproducibility is fragile. ConTree provides a branchable, VM-isolated sandbox
designed for rapid experimentation across thousands of SWE environments - over 7,000
are preloaded out of the box.

Purpose-built for SWE agent research
------------------------------------

- Thousands of ready-to-run SWE environments, so you
  can benchmark agents with minimal infrastructure setup.
- Git-like branching to try multiple patches or strategies in parallel from the
  same checkpoint, then keep only the winning branch: MCTS, Beam Search, rollbacks, value-function estimation.
- VM-level isolation with per-run metrics to keep generated code contained and
  experiments auditable.

Environment catalog
-------------------

Pick from the preloaded catalog rather than building or pulling terabytes of images yourself. What's included now:

- `SWE-bench Verified <https://huggingface.co/datasets/princeton-nlp/SWE-bench_Verified>`_
- `SWE-rebench <https://huggingface.co/datasets/nebius/SWE-rebench/viewer/default/filtered>`_

Integrations
------------

- `Out-of-the-box support <https://docs.contree.dev/sdk/integrations/mini-swe-agent.html>`_ for ``mini-swe-agent``: evaluate on thousands of preloaded SWE environments.
- Integrate your own agent via `ConTree SDK <https://docs.contree.dev/sdk/index.html>`_ or REST :doc:`api`.
- Connect via `MCP <https://docs.contree.dev/mcp/index.html>`_ to plug ConTree sandboxes directly into AI assistants that
  speak the Model Context Protocol.