Source: https://seerepo.com/compare/openai/codex/vs/vercel-labs/agent-desktop
Markdown source: https://seerepo.com/compare/openai/codex/vs/vercel-labs/agent-desktop.md
Description: codex and agent-desktop share 3 technologies: png, tracing, and tracing-subscriber. Both are written in Rust.

# codex vs agent-desktop

codex and agent-desktop share 3 technologies: png, tracing, and tracing-subscriber. Both are written in Rust.

- [openai/codex](https://seerepo.com/r/openai/codex): Lightweight coding agent that runs in your terminal
- [vercel-labs/agent-desktop](https://seerepo.com/r/vercel-labs/agent-desktop): Computer use CLI for AI agents

## What they share and what sets them apart

- In common: png, tracing, tracing-subscriber
- Only in codex: axum, lru, SQLite, Sentry, Tokio, tower, objc2-app-kit, objc2-foundation, Ratatui, crossterm, keyring, hmac, rmcp, image, pulldown-cmark, zip, schemars, Pydantic, log, tracing-appender
- Only in agent-desktop: x11rb, uiautomation, core-graphics

## Side by side

| | codex | agent-desktop |
|---|---|---|
| Size | 2,320,674 lines written, 4,216 code files | 412,850 lines written, 1,573 code files |
| Who writes it | 2.8% of its commits signed by AI. With Codex, Claude Code, and GitHub Copilot | 1% of its commits signed by AI. With Cursor and Claude Code |
| Activity | Last push November 4, 2026, 13,099 commits, 133,103 stars | Last push November 6, 2026, 301 commits, 18,108 stars |
| Tests | 2,448 test files. 51.6% of its code files have a test that imports them | 1,401 test files. 68% of its code files have a test that imports them |

- Size: Lines written by people and agents, without generated ones; code files, without tests.
- Who writes it: A commit counts as AI's when it carries its signature: a Co-Authored-By line, its author name or the merge of its branch.
- Activity: From GitHub and its history, on the day SeeRepo read it.
- Tests: The files that test the code, and the share of code files a test imports.

## Frequently asked questions

### Which is bigger, codex or agent-desktop?

codex, with 2,320,674 lines written; agent-desktop has 412,850.

The proof: [openai/codex's code at commit 021fccb](https://github.com/openai/codex/tree/021fccb046508938c4754982efccfb51d6e8356e)

### Which has more GitHub stars, codex or agent-desktop?

codex, with 133,103 stars; agent-desktop has 18,108.

The proof: [openai/codex on GitHub](https://github.com/openai/codex)

### Which has more tests, codex or agent-desktop?

codex, with 2,448 test files; agent-desktop has 1,401.

The proof: [openai/codex's code at commit 021fccb](https://github.com/openai/codex/tree/021fccb046508938c4754982efccfb51d6e8356e)

### Which was updated more recently, codex or agent-desktop?

agent-desktop: its last push was on November 6, 2026, and codex's on November 4, 2026.

The proof: [vercel-labs/agent-desktop's commits on GitHub](https://github.com/vercel-labs/agent-desktop/commits)

### Which has more commits signed by AI, codex or agent-desktop?

codex: an AI signed 2.8% of its commits, with Codex, Claude Code, and GitHub Copilot; in agent-desktop, 1%.

The proof: [the commits each tool signed in openai/codex](https://seerepo.com/r/openai/codex/ai)

## Sources

- openai/codex: Read on November 4, 2026, at commit 021fccb (https://github.com/openai/codex/tree/021fccb046508938c4754982efccfb51d6e8356e)
- vercel-labs/agent-desktop: Read on November 7, 2026, at commit b43ac4f (https://github.com/vercel-labs/agent-desktop/tree/b43ac4f6e892ef608f8851428613c4009f0e84c7)

## What is your repo built with?

Any public repo, with no account: https://seerepo.com/r/owner/repo/built, with the owner and name it has on GitHub.