Overview
Warp is an agentic development environment that grew out of a terminal. You can drive its own built-in coding agent, or bring your own CLI agent — Claude Code, Codex, Gemini CLI and others run inside it. The client codebase is open source and lives in the warpdotdev/warp repository, where the UI framework crates (`warpui_core` and `warpui`) are MIT-licensed and the rest of the code is AGPL v3; OpenAI is the founding sponsor of that open-source repository.
Alongside the desktop app, Warp ships the Warp Agent CLI: a standalone install of the same coding agent that runs in other terminals — Ghostty, iTerm 2, VS Code, Windows Terminal and macOS Terminal. It acts as a multiplexer across agent sessions, can operate on remote machines with no remote binary install, and supports multi-agent orchestration and handoff to cloud agents. Frontier and open-weight models are built in, and you can point it at a custom model router or bring your own API keys.
Development of the repository itself is unusually public: Warp Factories are code-defined agent pipelines that triage issues, write specs, implement changes and review pull requests, and build.warp.dev shows those agent sessions, in-flight features and top contributors live. Contribution is issue-first — maintainers apply `ready-to-spec` and `ready-to-implement` labels to signal which issues are open for design or code. Usage is commercial: the announcement lists a Warp subscription from $18/month including a $20 inference credit, ad-hoc credits from $10, or your own API key.
What it does
- A built-in coding agent plus first-class support for third-party CLI agents such as Claude Code, Codex and Gemini CLI
- Warp Agent CLI: a standalone install that runs in Ghostty, iTerm 2, VS Code, Windows Terminal and macOS Terminal
- Session multiplexing across agent sessions, and remote execution with no remote binary to install
- Frontier and open-weight models built in, with custom model routers and bring-your-own API keys
- Multi-agent orchestration and handoff of work to cloud agents
- Warp Factories — code-defined agent pipelines with evals and benchmarks that drive the repository's own issue-to-PR flow
Getting started
Install the Warp app for the full environment, or the standalone Agent CLI to use the agent inside a terminal you already run.
Install the Warp app
Download the build for your platform from warp.dev/download; the docs at docs.warp.dev carry the platform-specific instructions.
https://www.warp.dev/downloadInstall the Agent CLI on macOS or Linux
The standalone agent installs with a single script and then runs inside your existing terminal.
curl -fsSL https://app.warp.dev/download/agent-cli | bashInstall the Agent CLI on Windows
Use the PowerShell installer instead.
Invoke-RestMethod "https://app.warp.dev/download/agent-cli.ps1" | Invoke-ExpressionBuild from source
The client codebase is public. Clone it and follow CONTRIBUTING.md for the build and contribution flow; issues labelled ready-to-spec or ready-to-implement are the ones open for contributors.
git clone https://github.com/warpdotdev/warp.gitCommands and code are distilled from the project's own documentation — always check the official repo for the latest.
When to use it
- Run a coding agent in the terminal you already use, without switching editors
- Keep several agent sessions multiplexed in one place and hand long-running work off to cloud agents
- Work on a remote machine with an agent without installing anything on that machine
- Route agent work across frontier and open-weight models, or against your own keys and router
How Warp compares
Warp alongside other open-source autonomous coding agents tools AI/TLDR tracks, ranked by GitHub stars.
| Tool | Stars | What it does |
|---|---|---|
| opencode | ★ 205k | OpenCode is an open source AI coding agent that runs in your terminal, with built-in build and plan agents and an optional desktop app. |
| Claude Code | ★ 144k | Anthropic's agentic coding tool for the terminal, IDE and GitHub: it understands your codebase, executes routine tasks, explains code and handles git workflows from natural-language commands. |
| OpenAI Codex | ★ 122k | OpenAI's cloud and CLI coding agent that writes features, fixes bugs, and proposes code changes across a repo, running tasks in parallel. |
| Gemini CLI | ★ 107k | An open-source command-line AI agent from Google that connects your terminal to Gemini models for reading code, editing files, running shell commands, and searching the web. |
| Pi | ★ 102k | Minimal terminal coding agent harness with four built-in tools, extended through TypeScript extensions, skills and prompt templates instead of forks. |
| OpenHands | ★ 86.4k | An open-source AI software-development agent that plans tasks, edits files, runs commands, and tests code, usable from a terminal CLI, a local web GUI, or a Python SDK. |
| Warp | ★ 64.8k | An agentic development environment born out of the terminal, with a built-in multi-model coding agent and support for third-party CLI agents |
| Orca | ★ 62.7k | Open-source desktop workspace that runs Claude Code, Codex, OpenCode and other CLI coding agents in parallel, each in its own git worktree, with a mobile companion app. |