Refact
AI Agent that handles engineering tasks end-to-end: integrates with developers’ tools, plans, executes, and iterates until it achieves a successful result.
At a glance.
A compact read before the deeper capability notes and official setup links.
Core features.
Feature cards focus on what the tool helps users do, not generated setup commands.
Open-source, local-first AI coding assistant for IDE chat, autonomous agent workflows, and tool-powered development.
Refact runs a local Rust engine (refact-lsp) from your IDE and connects only to the model providers, local runtimes, and integrations you configure.
It brings together chat, codebase search, autonomous agents, browser automation, and tool integrations while keeping project state, indexes, trajectories, and task data on your machine.
Bring your own models: connect hosted providers, OpenAI-compatible endpoints, or local runtimes instead of relying on a bundled model service.
Deep codebase awareness: combine open files, selections, project tree, AST symbols, semantic search, git state, and previous work into useful model context.
Agent workflows inside the IDE: let the agent inspect files, edit code, run checks, use integrations, and report progress without leaving VS Code or JetBrains.
Extensible tools: use built-in tools, command-line integrations, browser automation, databases, code hosting integrations, and MCP servers.
It serves the chat UI, tracks open workspaces, exposes model capability and tool APIs, manages shutdown and background tasks, and keeps project state in local Refact directories.
Agent / Skill / MCP / Workflow fit.
This panel keeps technical format separate from the user-facing AI category.
Official setup path.
Generated install snippets are intentionally not mirrored here because they drift. The page links to source-owned setup docs instead.
Evidence and adoption notes.
These notes help a user decide whether to investigate the official project further.
Source repository last pushed at 2026-05-15T14:41:32Z.
Generated from source metadata; confirm operational details in the official project before adopting it.
Review the upstream license, maintenance activity, and issue history before using it in production.