langchain-ai/open-swe

An Open-Source Asynchronous Coding Agent

About the project

Open SWE is an open-source software factory that gives coding tasks to an asynchronous agent, which investigates repositories, changes code, validates the results, and delivers pull requests. Engineering teams use it to automate code changes, reviews, CI follow-up, and related repository work from dashboards and connected tools.

Latest release desktop-v0.3.10 — Open SWE Desktop v0.3.10 · 27 Oct 2026

6 files for Codex, OpenCode, Claude Code and GitHub Copilot: open-swe AGENTS.md, local-gh-dev, writing-pr-body-title-description, resolve_preview_conflict and 2 more — 2,256 tokens loaded in every session.

Open on GitHub langchain.com

▶ Introducing Open SWE: An Open-Source Asynchronous Coding Agent LangChain · about langchain-ai/open-swe · on YouTube →

Instructions

open-swe AGENTS.md AGENTS.md A 2,256 tok

Skills

local-gh-dev .claude/skills/local-gh-dev/SKILL.md A 72 tok
writing-pr-body-title-description .agents/skills/writing-pr-body-title-description/SKILL.md A 18 tok
dashboard-preview .claude/skills/dashboard-preview/SKILL.md A 55 tok

Commands

resolve_preview_conflict .github/prompts/resolve_preview_conflict.md A —
fix_preview_typecheck .github/prompts/fix_preview_typecheck.md A —

These files are langchain-ai/open-swe's own configuration — they tell Codex, OpenCode, Claude Code and GitHub Copilot how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.