Claw Code
ultraworkers/claw-codeReviewed by AI
Radar found strong risk signals in this project. It stays visible so you can read the evidence, but it is left out of rankings and recommendations.
What it does
Claw Code is a command-line AI coding assistant written in Rust. You type requests in a terminal and it talks to an AI model to work on code. The README says it is "not the serious production project here", calls it a "museum exhibit" run by AI agents with no human intervention, and points users to two other projects instead. It is for programmers curious about AI agent tools. The README says to install Rust and build it from source, though releases also include ready files for Windows, Mac and Linux. You need your own paid AI API key, such as Anthropic or OpenAI.
How easy it is to install
C · install 1 or 2 tools · Windows
- Windows C
- macOS C
- Linux C
All three platforms are the same work: the README says this repo is "build-from-source only", so you install Rust and Git, run `cargo build --workspace`, set an API key and use it from a terminal; bare release files (claw-windows-x64.exe, claw-macos-arm64, claw-linux-x64) also exist but the project does not document them as the install path.
Numbers on GitHub
- Stars
- 194,997on Oct 31, 2026
- Downloads
- 2,121496 this week
- Growth
- +9stars in 5 days
Stars over time
Facts
- Runs on
- Windows, macOS, Linux
- You need to install
- Rust
- To build from source
- Docker
- Disk space
- 15 MB–45 MB · from the installer
- Accounts or keys
- AI model provider key (Anthropic, OpenAI or a local OpenAI-compatible server)
- Cost
- Free, API costs extra · The software is free (MIT); it requires an API key, and Claude subscription login is not supported.
- License
- MIT (Permissive)
- Can replace
- Claude Code
- Last update
- Oct 28, 2026
Risk signals (9)
- Highpossible unlicensed reproduction
README.md: "This repository does not claim ownership of the original Claude Code source material" and "not affiliated with, endorsed by, or maintained by Anthropic"; the repository shape section describes "PARITY.md — Rust-port parity status and migration notes", and LICENSE puts the whole repository under MIT. A port built on another company's proprietary source material, redistributed under MIT, may be published without a licence for that material; whether any code was copied was not verified.
AI review
- Attentionunusual star rate
194971 stars in 211 days (~924 per day)
automatic check
- AttentionInstall command runs a script from the internet
Install command pipes a remote script into a shell: curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh
automatic check
- Attentionself described not production
README.md: "Claw Code is not the serious production project here. This repository is closer to a museum exhibit than a product pitch"; repository description: "developed and maintained with no human intervention".
AI review
- Attentionpossible reproduction of proprietary product
README.md: "This repository does not claim ownership of the original Claude Code source material" and "not affiliated with, endorsed by, or maintained by Anthropic"; PARITY.md is described as "Rust-port parity status".
AI review
- Infounusual fork count
107666 forks for 194971 stars (about 55%) and only 1 open issue, in a repository created 2026-03-31.
AI review
- Inforeadme redirects to other projects
README.md: "If you want to actually run work, start with LazyCodex or Gajae-Code", with large banners linking to those repositories.
AI review
- Infopackage name confusion
README.md: "`cargo install claw-code` installs the wrong thing. The `claw-code` crate on crates.io is a deprecated stub".
AI review
- Inforelease files contradict readme
The README says the repo is "build-from-source only", yet release v0.1.9 publishes ready binaries (claw-windows-x64.exe, claw-macos-arm64, claw-linux-x64) with no documented install instructions for them.
AI review
Evidence is shown in its original language.
What Radar could not check
- Whether the release files match this source code and work without building.
- Which files and project content are sent to the AI service while you use it.
- Real API cost for a typical coding session.
- Whether any of the code was copied from another company's product.
- How stable it is, given that AI agents maintain it without human review.