pocketcoder 1.0.1

AI-powered coding assistant optimized for local LLMs

PocketCoder

AI coding agent that runs with any LLM — local or cloud.

Stop paying $20/month for cloud AI coding tools. PocketCoder works with free local models like Qwen, Llama, or DeepSeek running on your own machine via Ollama. Or connect to OpenAI/Claude if you prefer.

Why PocketCoder?

Most AI coding assistants are locked to expensive cloud APIs. PocketCoder is different:

• Run locally for free — Use Ollama with open-source models (7B-14B work great)
• Session memory — Agent remembers what you're working on across requests
• Project awareness — RepoMap shows your codebase structure to the LLM
• Multi-step tasks — TODO tracking, automatic planning, checkpoint progress
• Any LLM provider — Ollama, OpenAI, Anthropic, vLLM, LM Studio, or any OpenAI-compatible API

Installation

pip install pocketcoder

Quick Start

# Start in your project directory
pocketcoder

# First run opens setup wizard to configure your LLM

With Ollama (free, local)

# Install Ollama: https://ollama.com
ollama pull qwen2.5-coder:7b
pocketcoder

With OpenAI

export OPENAI_API_KEY="sk-..."
pocketcoder --provider openai --model gpt-4o

How It Works

PocketCoder is a coding agent, not a chatbot. When you give it a task:

1. Understands context — Reads your files, understands project structure
2. Plans the work — Breaks complex tasks into steps
3. Executes tools — Reads files, writes code, runs commands
4. Validates results — Checks its own work before marking done

> add authentication to the API

[TODO] Read existing API structure
[TODO] Create auth middleware
[TODO] Add login/logout endpoints
[TODO] Update existing routes

Reading api/routes.py...
Creating api/auth.py...
Done: Added JWT authentication with login/logout endpoints

Commands

Command	Description
/help	Show available commands
/model	Switch LLM model
/add <file>	Add file to context
/drop <file>	Remove file from context
/files	List files in context
/undo	Undo last file change
/clear	Clear conversation history
/quit	Exit

Supported Providers

Provider	Type	Setup
Ollama	Local	ollama serve then ollama pull model
OpenAI	Cloud	Set OPENAI_API_KEY
Anthropic	Cloud	Set ANTHROPIC_API_KEY
vLLM	Local/Cloud	Any OpenAI-compatible endpoint
LM Studio	Local	Run server, point to localhost

Configuration

Config stored in ~/.pocketcoder/config.yaml:

provider:
  type: ollama
  base_url: http://localhost:11434
  default_model: qwen2.5-coder:7b

thinking:
  mode: smart           # smart | always | never
  show_reasoning: true  # show agent's thinking process

Recommended Models

For local use with 16GB+ RAM:

Model	Size	Speed	Quality
qwen2.5-coder:7b	4.7GB	Fast	Good
qwen2.5-coder:14b	9GB	Medium	Better
deepseek-coder:6.7b	4GB	Fast	Good
codellama:13b	7GB	Medium	Good

Project Structure

PocketCoder stores session data in .pocketcoder/ directory (like .git/):

.pocketcoder/
  project_context.json  # Current session state
  episodes.jsonl        # Conversation history
  memory.json           # Long-term facts

Development

git clone https://github.com/Chashchin-Dmitry/pocketcoder.git
cd pocketcoder
pip install -e ".[dev]"
pytest

Support the Project

If PocketCoder saves you money on AI subscriptions, consider supporting development:

Network	Address
ETH / USDT (ERC-20)	0xdF5e04d590d44603FDAdDb9f311b9dF7E5dE797c
BTC	bc1q3q25vw4jm8v4xe2g6uezq35q2uyn5jt6e00mj9
USDT (TRC-20)	TQj3X5nFQWqPEmRUWNFPjkaRUUFLxmCdok
SOL	5s5uP66VmnLMSApjq8ro639tXvSp59XEwQittzxF64mF

License

MIT License