npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

octonoesis

v0.1.1

Published

An open-source, lightweight terminal coding agent built on Bun + Ink

Readme

Octonoesis 🐙

An open-source, lightweight, and lightning-fast terminal coding agent designed to read code, search directories, edit files with unified diff approvals, run tests, and automatically execute commands to fulfill natural-language tasks.

Built entirely in TypeScript on the Bun runtime using Ink for a rich, responsive Terminal User Interface (TUI).


Architecture Flow

       ┌────────────────────────┐
       │ CLI input / TUI Prompt │
       └───────────┬────────────┘
                   │
                   ▼
       ┌────────────────────────┐
       │ buildSystemMessages()  │ ◄── OS, shell, CWD, git status, time
       └───────────┬────────────┘
                   │
                   ▼
       ┌────────────────────────┐
       │   LLMProvider Stream   │ ◄── Pinned Model & System Prompt
       └───────────┬────────────┘
                   ├─────────────────────────┐
                   ▼ (text_delta)            ▼ (tool_use)
       ┌────────────────────────┐    ┌────────────────────────┐
       │   Ink UI Text Stream   │    │  Zod Input Validation  │
       └────────────────────────┘    └───────────┬────────────┘
                                                 │
                                                 ▼
                                     ┌────────────────────────┐
                                     │ Permission Interceptor │
                                     └───────────┬────────────┘
                                                 ├───────────────────────┐
                                                 ▼ (Approved / Read-only)▼ (Denied)
                                     ┌────────────────────────┐ ┌────────┴────────┐
                                     │     Tool Execution     │ │  tool_result    │
                                     │ (Read, Edit, Bash,...) │ │  "user_denied" │
                                     └───────────┬────────────┘ └────────┬────────┘
                                                 │                       │
                                                 └──────────┬────────────┘
                                                            │
                                                            ▼
                                                Append to message history
                                                & loop back to Provider

Features

  • Standardized Tool System: Executes tools serially with strict parameter validation (Zod) and sandbox safety boundaries (paths must remain within the repository root).
  • Interactive Permission UI: Prompts for [y] yes / [n] no / [a] always on modifying actions (like Edit or Bash), complete with colorful unified diff previews.
  • Robust Cancellation & Retry: Interrupt running processes and model streams cleanly with Ctrl+C. Handles rate limits (429) and server drops (5xx) with exponential backoff and jitter.
  • Provider Abstraction: First-class tested support for Anthropic Claude, OpenAI GPT-4o, and DeepSeek, with easy endpoint configuration.
  • Dynamic Context Suffix: Computes runtime environment status (OS, Shell, Git porcelain status, Time, Token usage) to ground LLM context dynamically.

Installation

Prerequisites

  • Bun (version >= 1.2.0 is required). To install Bun:
    curl -fsSL https://bun.sh/install | bash
  • ripgrep (optional fallback; recommended if prebuilt @vscode/ripgrep is blocked by system policies):
    # macOS
    brew install ripgrep
    # Debian/Ubuntu
    sudo apt-get install ripgrep

Install Globally

bun install -g octonoesis

Quickstart (5 Minutes)

  1. Set your API Key in your environment:

    # For Anthropic (default)
    export ANTHROPIC_API_KEY="your-api-key"
    
    # For OpenAI
    export LLM_PROVIDER="openai"
    export OPENAI_API_KEY="your-api-key"
  2. Run the agent in one of two modes:

    • Interactive TUI Mode:

      octonoesis

      This launches a full terminal dashboard showing the LLM conversation stream on the left, and a live in-memory TODO status panel on the right.

    • One-shot Mode:

      octonoesis "Fix the spelling mistake in src/utils/errors.ts"

      This streams the solution directly to standard output and exits.


Documentation

To learn more about the technical details, architecture, and design decisions of the project:


License

This project is licensed under the MIT License. See LICENSE for details.