Skip to main content

Production-grade TUI for AI-assisted development with RLM context management

Project description

PinkyClawd

╭──────────────────────────────────────────────────────────╮
│                                                          │
│   ██████╗ ██╗███╗   ██╗██╗  ██╗██╗   ██╗                │
│   ██╔══██╗██║████╗  ██║██║ ██╔╝╚██╗ ██╔╝                │
│   ██████╔╝██║██╔██╗ ██║█████╔╝  ╚████╔╝                 │
│   ██╔═══╝ ██║██║╚██╗██║██╔═██╗   ╚██╔╝                  │
│   ██║     ██║██║ ╚████║██║  ██╗   ██║                   │
│   ╚═╝     ╚═╝╚═╝  ╚═══╝╚═╝  ╚═╝   ╚═╝                   │
│                                                          │
│   ██████╗██╗      █████╗ ██╗    ██╗██████╗              │
│   ██╔════╝██║     ██╔══██╗██║    ██║██╔══██╗            │
│   ██║     ██║     ███████║██║ █╗ ██║██║  ██║            │
│   ██║     ██║     ██╔══██║██║███╗██║██║  ██║            │
│   ╚██████╗███████╗██║  ██║╚███╔███╔╝██████╔╝            │
│    ╚═════╝╚══════╝╚═╝  ╚═╝ ╚══╝╚══╝ ╚═════╝             │
│                                                          │
│           ◢◤ AI-Powered Development ◥◣                  │
╰──────────────────────────────────────────────────────────╯

AI-powered development tool with TUI interface and Recursive Language Model (RLM) context management for unlimited conversation context.

Features

  • RLM Context Management: Automatic context archival and retrieval enables conversations beyond context window limits
  • Multi-Provider Support: Anthropic (Claude), OpenAI (GPT-4), and more
  • Rich TUI Interface: Full-featured terminal UI with Textual
  • 14 Built-in Tools: bash, read, write, edit, glob, grep, todo, webfetch, memory, and more
  • Session Management: Fork, export, and manage conversation sessions
  • Slash Commands: /context, /search, /compact, /models, /agents

Installation

via npm (recommended)

npm install -g pinkyclawd

via pip

pip install pinkyclawd

From source

git clone https://github.com/tekcin/PinkyClawd.git
cd PinkyClawd
pip install -e .

Requirements

  • Python 3.11+
  • Node.js 16+ (for npm installation)

Usage

# Start the TUI
pinkyclawd

# Start with a specific project
pinkyclawd /path/to/project

# Continue last session
pinkyclawd --continue

# Run a single command (non-interactive)
pinkyclawd run "explain this codebase"

# List available models
pinkyclawd models

Configuration

Create ~/.config/pinkyclawd/pinkyclawd.json:

{
  "model": "anthropic/claude-sonnet-4",
  "theme": "pinkyclawd",
  "rlm": {
    "enabled": true,
    "threshold_ratio": 0.33,
    "auto_retrieve": true
  }
}

Environment Variables

export ANTHROPIC_API_KEY=sk-ant-...
export OPENAI_API_KEY=sk-...

RLM (Recursive Language Model)

RLM enables unlimited context by automatically:

  1. Tracking token usage across the conversation
  2. Archiving older context when threshold is reached (default: 33%)
  3. Retrieving relevant archived context for new queries
  4. Injecting context into the conversation seamlessly

Manual RLM Commands

  • /context - Show current token usage
  • /search <query> - Search archived context
  • /compact - Manually trigger context archival

RLM Tools

  • memory - Search and retrieve archived context
  • rlm_query - Execute Python queries on archived context

License

MIT

Author

Michael Thornton (tekcin@yahoo.com)

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

pinkyclawd-0.2.0.tar.gz (238.2 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

pinkyclawd-0.2.0-py3-none-any.whl (310.4 kB view details)

Uploaded Python 3

File details

Details for the file pinkyclawd-0.2.0.tar.gz.

File metadata

  • Download URL: pinkyclawd-0.2.0.tar.gz
  • Upload date:
  • Size: 238.2 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.14

File hashes

Hashes for pinkyclawd-0.2.0.tar.gz
Algorithm Hash digest
SHA256 df76a2c1cb93c61c6852d85c98490641985419cf67048513126a681c5c52be5d
MD5 7b4c2427513c9abd1ea73811cf3928e9
BLAKE2b-256 78057a8d833fe2cb306af179085be62618bfc4d98e83e57d23d6a2732226ca6c

See more details on using hashes here.

File details

Details for the file pinkyclawd-0.2.0-py3-none-any.whl.

File metadata

  • Download URL: pinkyclawd-0.2.0-py3-none-any.whl
  • Upload date:
  • Size: 310.4 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.14

File hashes

Hashes for pinkyclawd-0.2.0-py3-none-any.whl
Algorithm Hash digest
SHA256 0534635410f6706adc12ec17d86fcde95ea9a7535e3a9fcd6416430b3c1f9a19
MD5 373782bebe33094a57b54d253c85349c
BLAKE2b-256 621acae486d9550134edf95325c93ec003c646cd290b8358584894a0d0144b58

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page