Coding agent · Cloud coding agent (self-hosted)

Open SWE

Open SWE is a Cloud coding agent (self-hosted) in the 2026 agents.dancingteeth catalog. Teams who want their own Stripe/Ramp-style internal coding agent: MIT code they can fork, triggered from Slack, GitHub, Linear or a dashboard, with sandboxed PRs, PR reviews and CI watching.

Last updated

Verified

Best for

Teams who want their own Stripe/Ramp-style internal coding agent: MIT code they can fork, triggered from Slack, GitHub, Linear or a dashboard, with sandboxed PRs, PR reviews and CI watching.

Skip if

You want a hosted agent you can sign up for today, have no Managed Deep Agents beta access and no time to fork, need repo-level isolation inside one deployment, or need a stable API.

Who it fits

  • Teams who want their own Stripe/Ramp-style internal coding agent: MIT code they can fork, triggered from Slack, GitHub, Linear or a dashboard, with sandboxed PRs, PR reviews and CI watching.
  • Builders who start in the browser without a local IDE setup
  • Terminal-first engineers who live in the shell and git
  • People who prefer a dedicated desktop coding surface
Experimental

Pros

LangChain's open-source software factory on Deep Agents + LangGraph: investigates a repo, implements in an isolated per-thread sandbox, validates and opens a PR; reviews PRs and learns repo review preferences; watches CI on opted-in PRs (/baby-sit), scheduled tasks, read-only PR chat, subagents; start work from the web dashboard, GitHub issues/PR comments, Slack or Linear; `oswe` CLI and experimental desktop app can use your local checkout; pick models, sandbox provider (LangSmith default; Daytona, Runloop, E2B, Modal), skills and MCP per workspace.

Cons

Since Oct 16, 2026 every thread runs on LangChain Managed Deep Agents (private beta), so deploying it as-is needs beta access; otherwise you fork and adapt it; no hosted public app; LangChain says breaking changes are expected and it does not accept issues or contributions; sandboxes get installation-wide GitHub App access, so it is meant for one trusted team per deployment; you run the LangSmith, GitHub App and Slack setup yourself.

Facts

Locality:
Hybrid (local + cloud)
Surfaces:
Web, CLI, Desktop
Maturity:
Experimental
Open source:
Source available
MCP:
Model Context Protocol support
Verified:
Oct 2026

Agent standards & memory

AGENTS.md: SupportedSKILL.md: SupportedRules / memory: Repo-root AGENTS.md appended to the system prompt; admin repo instructions; bundled, organization and per-user SKILL.md skills; per-thread persistent sandbox and LangGraph thread state.Free and open source (MIT); you pay for your own model keys or LangSmith Gateway, LangSmith/sandbox usage and Managed Deep Agents (private beta, no published price).

Compare one neighbor first

Vendor pages sell hard. Read one alternative on this site, then leave for the official link if the fit still holds.

Alternatives

Related tools

Frequently asked questions

What is Open SWE?

Open SWE is a Cloud coding agent (self-hosted) covered in the agents.dancingteeth 2026 catalog. LangChain's open-source software factory on Deep Agents + LangGraph: investigates a repo, implements in an isolated per-thread sandbox, validates and opens a PR; reviews PRs and learns repo review preferences; watches CI on opted-in PRs (/baby-sit), scheduled tasks, read-only PR chat, subagents; start work from the web dashboard, GitHub issues/PR comments, Slack or Linear; `oswe` CLI and experimental desktop app can use your local checkout; pick models, sandbox provider (LangSmith default; Daytona, Runloop, E2B, Modal), skills and MCP per workspace.

Who is Open SWE best for?

Teams who want their own Stripe/Ramp-style internal coding agent: MIT code they can fork, triggered from Slack, GitHub, Linear or a dashboard, with sandboxed PRs, PR reviews and CI watching.

When should I skip Open SWE?

You want a hosted agent you can sign up for today, have no Managed Deep Agents beta access and no time to fork, need repo-level isolation inside one deployment, or need a stable API.

What are the main trade-offs of Open SWE?

Pros: LangChain's open-source software factory on Deep Agents + LangGraph: investigates a repo, implements in an isolated per-thread sandbox, validates and opens a PR; reviews PRs and learns repo review preferences; watches CI on opted-in PRs (/baby-sit), scheduled tasks, read-only PR chat, subagents; start work from the web dashboard, GitHub issues/PR comments, Slack or Linear; `oswe` CLI and experimental desktop app can use your local checkout; pick models, sandbox provider (LangSmith default; Daytona, Runloop, E2B, Modal), skills and MCP per workspace. Cons: Since Oct 16, 2026 every thread runs on LangChain Managed Deep Agents (private beta), so deploying it as-is needs beta access; otherwise you fork and adapt it; no hosted public app; LangChain says breaking changes are expected and it does not accept issues or contributions; sandboxes get installation-wide GitHub App access, so it is meant for one trusted team per deployment; you run the LangSmith, GitHub App and Slack setup yourself.

Does Open SWE support AGENTS.md, SKILL.md, or MCP?

This card lists AGENTS.md as Supported, SKILL.md as Supported, and rules/memory as Repo-root AGENTS.md appended to the system prompt; admin repo instructions; bundled, organization and per-user SKILL.md skills; per-thread persistent sandbox and LangGraph thread state.. Free options: Free and open source (MIT); you pay for your own model keys or LangSmith Gateway, LangSmith/sandbox usage and Managed Deep Agents (private beta, no published price)..

Someone asks you about vibe-coding software? Share this site with them — we are adding more useful data regularly. agents.dancingteeth.net

← Back to full tool directory