Metadata-Version: 2.4
Name: arcmesh
Version: 0.1.1
Summary: One command. Your codebase, AI-ready.
License: MIT
Project-URL: Homepage, https://github.com/arcmesh-labs/arcmesh-cli
Project-URL: Bug Tracker, https://github.com/arcmesh-labs/arcmesh-cli/issues
Keywords: mcp,claude,ai,developer-tools,llm
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Requires-Python: >=3.10
Description-Content-Type: text/markdown
Requires-Dist: click>=8.1
Requires-Dist: rich>=13.0
Requires-Dist: questionary>=2.0

# ArcMesh CLI

**One command. Your codebase, AI-ready.**

```bash
pip install arcmesh
cd your-project
mcp setup
```

Restart Claude Desktop. Done.

---

## What it does

ArcMesh creates a local MCP server for your project and connects it to Claude Desktop automatically.

No config files. No decisions. No path issues.

```bash
✅ AI workspace ready

Project:    your-project
MCP server: .mcp/server.py

Next steps:
  1. Restart Claude Desktop
  2. Open this folder in Claude Desktop

Try asking:
  "Explain this repo"
  "Find all API endpoints"
  "Where is authentication handled?"
```

---

## Requirements

- Python 3.10+
- Claude Desktop
- Mac, Windows, or WSL

---

## Advanced

Most people never need this. But if you want full control:

| Command | What it does |
|---|---|
| `mcp init` | Initialize config without generating a server |
| `mcp add` | Add a custom MCP server (wizard or inline) |
| `mcp sync` | Push local config to Claude Desktop |
| `mcp status` | Show configured servers |
| `mcp remove` | Remove a server |
| `mcp unwrap` | Import WSL-wrapped servers back to local config |

Run `mcp --help` to see everything.

---

## Philosophy

There should be one default way to make a codebase AI-ready.

One command. No decisions.

---

## License

MIT
