Claw Code Agent

A Python reimplementation of the Claude Code agent architecture — local models, full control.

---

📖 About

This repository reimplements the Claude Code npm agent architecture entirely in Python, designed to run with local open-source models via an OpenAI-compatible API server.

Built on the public porting workspace from instructkr/claw-code, the active development lives at HarnessLab/claw-code-agent.

Goal: Not to ship the original npm source, but to reimplement the full agent flow in Python — prompt assembly, context building, slash commands, tool calling, session persistence, and local model execution.

---

✨ Key Features

Feature	Description
🤖 Agent Loop	Full agentic coding loop with tool calling and iterative reasoning
🧰 Core Tools	File read / write / edit, glob search, grep search, shell execution
💬 Slash Commands	Local commands: /help, /context, /tools, /memory, /status, /model, and more
🧠 Context Engine	Automatic context building with CLAUDE.md discovery and usage reporting
🔄 Session Persistence	Save and resume agent sessions across runs
🔐 Permission System	Granular control: --allow-write, --allow-shell, --unsafe
🏗️ OpenAI-Compatible	Works with any OpenAI-compatible model server (vLLM, Ollama, etc.)
🐉 Qwen3-Coder	First-class support for Qwen3-Coder-30B-A3B-Instruct via vLLM

---

📋 Roadmap

Done

• Python CLI agent loop
• OpenAI-compatible local model backend
• Qwen3-Coder support through vLLM with qwen3_xml tool parser
• Core tools: list_dir, read_file, write_file, edit_file, glob_search, grep_search, bash
• Context building and /context-style usage reporting
• Slash commands: /help, /context, /context-raw, /prompt, /permissions, /model, /tools, /memory, /status, /clear
• Session persistence and agent-resume flow
• Permission system (read-only, write, shell, unsafe tiers)
• Unit tests for the Python runtime
• pyproject.toml packaging with setuptools

In Progress

• Full plugin and MCP server parity
• Complete slash-command parity with npm runtime
• Full interactive REPL / session persistence parity
• Tokenizer-accurate context accounting
• Remote runtime modes (SSH, teleport, deep-link)
• Voice and VIM modes
• Hooks system
• Cost tracking and budget limits

---

🏗️ Architecture

claw-code/
├── README.md
├── pyproject.toml
├── .gitignore
├── images/
│   └── logo.png
├── src/                          # Python implementation
│   ├── main.py                   # CLI entry point & argument parsing
│   ├── agent_runtime.py          # Core agent loop (LocalCodingAgent)
│   ├── agent_tools.py            # Tool definitions & execution engine
│   ├── agent_prompting.py        # System prompt assembly
│   ├── agent_context.py          # Context building & CLAUDE.md discovery
│   ├── agent_context_usage.py    # Context usage estimation & reporting
│   ├── agent_session.py          # Session state management
│   ├── agent_slash_commands.py   # Local slash command processing
│   ├── agent_types.py            # Shared dataclasses & type definitions
│   ├── openai_compat.py          # OpenAI-compatible API client
│   ├── session_store.py          # Session serialization & persistence
│   ├── permissions.py            # Tool permission filtering
│   ├── tools.py                  # Mirrored tool inventory
│   ├── commands.py               # Mirrored command inventory
│   ├── ...                       # 75+ modules across 30+ packages
│   ├── plugins/                  # Plugin subsystem (WIP)
│   ├── hooks/                    # Hook system (WIP)
│   ├── remote/                   # Remote runtime modes (WIP)
│   ├── voice/                    # Voice mode (WIP)
│   └── vim/                      # VIM mode (WIP)
└── tests/                        # Unit tests
    ├── test_agent_runtime.py
    ├── test_agent_context.py
    ├── test_agent_context_usage.py
    ├── test_agent_prompting.py
    ├── test_agent_slash_commands.py
    └── test_porting_workspace.py

---

📦 Requirements

Requirement	Details
🐍 Python	3.10 or higher
🖥️ Model Server	Any OpenAI-compatible server (vLLM recommended)
🧠 Model	Qwen/Qwen3-Coder-30B-A3B-Instruct (recommended)

---

🚀 Quick Start

1. Start vLLM with Qwen3-Coder

vLLM must be started with automatic tool choice enabled. Use the qwen3_xml parser for Qwen3-Coder tool calling:

python -m vllm.entrypoints.openai.api_server \
  --model Qwen/Qwen3-Coder-30B-A3B-Instruct \
  --host 127.0.0.1 \
  --port 8000 \
  --enable-auto-tool-choice \
  --tool-call-parser qwen3_xml

Verify the server is running:

curl http://127.0.0.1:8000/v1/models

📚 References: vLLM Tool Calling Docs · OpenAI-Compatible Server

2. Configure Environment

export OPENAI_BASE_URL=http://127.0.0.1:8000/v1
export OPENAI_API_KEY=local-token
export OPENAI_MODEL=Qwen/Qwen3-Coder-30B-A3B-Instruct

3. Run the Agent

# Read-only question
python3 -m src.main agent \
  "Read src/agent_runtime.py and summarize how the loop works." \
  --cwd .

# Write-enabled task
python3 -m src.main agent \
  "Create TEST_QWEN_AGENT.md with one line: test ok" \
  --cwd . --allow-write

# Shell-enabled task
python3 -m src.main agent \
  "Run pwd and ls src, then summarize the result." \
  --cwd . --allow-shell

---

🛠️ Usage

Agent Commands

Command	Description
agent <prompt>	Run the agent with a prompt
agent-prompt	Show the assembled system prompt
agent-context	Show estimated context usage
agent-context-raw	Show the raw context snapshot
agent-resume <id> <prompt>	Resume a saved session

CLI Flags

Flag	Description
--cwd <path>	Set the workspace directory
--model <name>	Override the model name
--base-url <url>	Override the API base URL
--allow-write	Allow the agent to modify files
--allow-shell	Allow the agent to execute shell commands
--unsafe	Allow destructive shell operations
--show-transcript	Print the full message transcript
--system-prompt <text>	Set a custom system prompt
--append-system-prompt <text>	Append to the system prompt
--add-dir <path>	Add extra directories to context

Slash Commands

These are handled locally before the model loop:

Command	Aliases	Description
/help	/commands	Show built-in slash commands
/context	/usage	Show estimated session context usage
/context-raw	/env	Show raw environment & context snapshot
/prompt	/system-prompt	Render the effective system prompt
/permissions	—	Show active tool permission mode
/model	—	Show or update the active model
/tools	—	List registered tools with permission status
/memory	—	Show loaded CLAUDE.md memory bundle
/status	/session	Show runtime/session status summary
/clear	—	Clear ephemeral runtime state

python3 -m src.main agent "/help"
python3 -m src.main agent "/context" --cwd .
python3 -m src.main agent "/tools" --cwd .
python3 -m src.main agent "/status" --cwd .

Utility Commands

python3 -m src.main summary            # Workspace summary
python3 -m src.main manifest           # Workspace manifest
python3 -m src.main commands --limit 10 # Command inventory
python3 -m src.main tools --limit 10    # Tool inventory

---

🔧 Built-in Tools

The agent has access to 7 core tools:

Tool	Description	Permission
list_dir	List files and directories	🟢 Always
read_file	Read file contents (with line ranges)	🟢 Always
write_file	Write or create files	🟡 --allow-write
edit_file	Edit files via exact string matching	🟡 --allow-write
glob_search	Find files by glob pattern	🟢 Always
grep_search	Search file contents by regex	🟢 Always
bash	Execute shell commands	🔴 --allow-shell

---

🔄 Session Persistence

Each agent run automatically saves a resumable session:

session_id=4f2c8c6f9c0e4d7c9c7b1b2a3d4e5f67
session_path=.port_sessions/agent/4f2c8c6f...

Resume a previous session:

python3 -m src.main agent-resume \
  4f2c8c6f9c0e4d7c9c7b1b2a3d4e5f67 \
  "Continue the previous task and finish the missing parts."

Inspect saved sessions:

ls -lt .port_sessions/agent

Note: Run agent-resume from the same claw-code/ directory where the session was created. A resumed session continues from the saved transcript, not from scratch.

---

🧪 Testing

Run the full test suite:

python3 -m unittest discover -s tests -v

Smoke tests:

python3 -m src.main agent "/help"
python3 -m src.main agent-context --cwd .
python3 -m src.main agent \
  "Read src/agent_session.py and summarize the message flow." \
  --cwd .

---

🔐 Permission Model

Claw Code Agent uses a tiered permission system to keep the agent safe by default:

Tier	Capability	Flag Required
Read-only	List, read, glob, grep	None (default)
Write	+ file creation and editing	--allow-write
Shell	+ shell command execution	--allow-shell
Unsafe	+ destructive shell operations	--unsafe

---

⚠️ Disclaimer

• This repository is a Python reimplementation inspired by the Claude Code npm architecture.
• It does not ship the original npm source.
• It is not affiliated with or endorsed by Anthropic.

---

_{Built with 🐍 Python · Powered by 🐉 HarnessLab Team.}