Skillget

claudecode

agent
Security Audit
Warn
Health Warn
  • No license — Repository has no license file
  • Description — Repository has a description
  • Active repo — Last push 0 days ago
  • Community trust — 116 GitHub stars
Code Pass
  • Code scan — Scanned 12 files during light audit, no dangerous patterns found
Permissions Pass
  • Permissions — No dangerous permissions requested
Purpose
This tool is a local coding-agent CLI written in Rust that replicates the functionality of Anthropic's Claude Code. It provides an interactive agent shell, one-shot prompts, and workspace-aware tools to assist developers directly in their terminal.

Security Assessment
Overall Risk: Medium. The light audit found no hardcoded secrets, dangerous code patterns, or excessive permission requests. However, as an AI coding agent, the tool inherently processes sensitive data, reads workspace files, and executes shell commands to function properly. It also makes external network requests to AI provider APIs (Anthropic, xAI) using user-provided API keys. Users should be aware that their prompts and local files will be sent to these third-party endpoints.

Quality Assessment
The project shows promising early community traction with 116 GitHub stars and is actively maintained, with the most recent push occurring today. A major concern for enterprise or collaborative use is the complete lack of a license file; without one, default copyright laws strictly apply, meaning the code cannot be legally modified or distributed by others. The repository structure is clean, the instructions are comprehensive, and the codebase is straightforward to compile and run.

Verdict
Use with caution.
SUMMARY

Open Source ClaudeCode Leaked. Same functionalities as Claude Code by Anthropic. Written in Rust, Ready to use immediately for free.

README.md

ClaudeCode

ClaudeCode is an open source Anthropic Claude Code with Rust implementation. It is a local coding-agent CLI that provides an interactive agent shell, one-shot prompts, workspace-aware tools, plugin support, and resumeable sessions.

View Count

Star History Chart

Start here

  1. Install Rust stable and Cargo.
  2. Set up credentials for the model provider you want to use.
  3. Run the CLI from the rust/ workspace.

First run

From the repository root:

cd rust

Build the CLI:

cargo build --release -p claw-cli

Start the interactive shell:

cargo run --bin claw -- --help
cargo run --bin claw --

Run a single prompt:

cargo run --bin claw -- prompt "summarize this workspace"

Install it locally if you want the binary on your PATH:

cargo install --path crates/claw-cli --locked

Authentication

Use whichever provider you have access to.

Anthropic-compatible models:

export ANTHROPIC_API_KEY="..."
export ANTHROPIC_BASE_URL="https://api.anthropic.com"

Grok models:

export XAI_API_KEY="..."
export XAI_BASE_URL="https://api.x.ai"

You can also authenticate through the CLI:

cargo run --bin claw -- login

Common commands

cargo run --bin claw -- prompt "review the latest changes"
cargo run --bin claw -- init
cargo run --bin claw -- logout
cargo run --bin claw -- --resume session.json /status

Run claw --help for the full command list, including agents, skills, system-prompt output, and slash-command flows.

Repository layout

.
├── rust/            # Active Rust workspace and CLI/runtime implementation
├── src/             # Python porting workspace and support code
├── tests/           # Python verification
├── CLAW.md          # Repo-specific working notes
└── README.md        # This guide

Development

Use the Rust workspace for local verification:

cd rust
cargo fmt
cargo clippy --workspace --all-targets -- -D warnings
cargo test --workspace

If you change the Python workspace in src/, keep the matching tests in tests/ updated too.

Notes

  • This repo is an open source Claude Code-style Rust implementation, not the original Claude Code source.
  • CLAW.md contains the workflow guidance for contributors working in this tree.

Reviews (0)

No results found