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

binote

v0.3.4

Published

Bidirectional-linked project memory system for AI, exposed as MCP server

Readme

Binote

Bidirectional-linked project memory for AI. Mirrors your codebase into .binote/ with [[wiki-style links]], giving AI agents structured context about your project's architecture, design decisions, and file relationships.

Why

AI tools read code but lack project context — why a file exists, how modules relate, what design tradeoffs were made. Binote fills this gap with a local knowledge graph that lives alongside your code.

How it works

myproject/
├── src/
│   ├── index.ts
│   └── utils/
│       └── helpers.ts
├── .binote/                  ← generated
│   ├── _dir.md             ← root overview
│   ├── _notes/             ← standalone docs (architecture, design, etc.)
│   │   └── architecture.md
│   ├── _index.json         ← link index cache
│   ├── src/
│   │   ├── _dir.md         ← directory overview
│   │   ├── index.ts.md     ← note for index.ts
│   │   └── utils/
│   │       ├── _dir.md
│   │       └── helpers.ts.md

Notes use [[bidirectional links]] — link to any file or note by path:

# src/index.ts

Entry point. Orchestrates [[src/utils/helpers.ts]] and [[_notes/architecture.md]].

Use depth=1 in read_note to expand all linked and backlinked notes in one call.

Install

npm install -g binote

Setup

Claude Plugin (recommended)

/plugin marketplace add Guxi11/binote
/plugin install binote

This gives you both MCP tools and slash commands.

Claude MCP only

claude mcp add binote -- binote

MCP tools only — no slash commands.

Local dev

claude --plugin-dir /path/to/binote

See CONTRIBUTING.md for dev setup.

Manual MCP config

Add to your MCP config (.mcp.json, claude_desktop_config.json, etc.):

{
  "mcpServers": {
    "binote": {
      "command": "binote"
    }
  }
}

Slash Commands

| Command | Description | |---------|-------------| | /binote:mode | Activate binote-first mode — reads .binote/ notes before source files. When you @file, the binote note is read first. | | /binote:save | Save current session's learnings (design decisions, discoveries) into .binote/ notes. | | /binote:rule | Emit a CLAUDE.md snippet for always-on binote-first behavior. |

Slash commands require plugin install. See Setup above.

MCP Tools

| Tool | Description | |------|-------------| | init | Scan project and generate .binote/ skeleton | | read_note | Read notes — accepts exact paths or [[link]] targets. depth=1+ recursively expands linked and backlinked notes (cycle-safe) | | write_note | Create or update a note with [[links]] | | search | Full-text search across all notes | | sync | Detect file renames/deletes, mark orphaned notes | | rebuild_index | Rebuild _index.json link graph from all notes | | list_notes | List all notes in .binote/ |

CLI

binote init     [projectRoot]                  # Initialize .binote/
binote list     [projectRoot]                  # List all notes
binote read     <notePath> [projectRoot]       # Read a note (fuzzy resolve)
binote write    <notePath> <content> [root]    # Write a note
binote links    <notePath> [projectRoot]       # Query forward/backlinks
binote search   <query> [projectRoot]          # Full-text search
binote resolve  <target> [projectRoot]         # Resolve a [[link]] target
binote dangling [projectRoot]                  # List all unresolved [[links]]
binote sync     [projectRoot]                  # Detect changes, mark orphans

No arguments starts the MCP server (stdio transport).

Workflow

  1. init — scaffold .binote/ with empty notes mirroring your source files
  2. /binote:mode — activate binote-first mode in any conversation
  3. Work normally — Claude reads binote notes before source files for faster context
  4. /binote:save — capture session learnings back into notes
  5. sync — after refactoring, detect structural changes and mark orphans

Git

Add _index.json to your .gitignore — it's a derived cache rebuilt automatically on demand:

.binote/_index.json

Design

  • Plain markdown — human-readable, editable, git-friendly
  • No database — link index is a JSON cache rebuilt on demand
  • Mixed authorship — AI and human can both read and write notes
  • Non-code files filtered — JSON, YAML, images, etc. are not mirrored
  • Zero config — just init and go

License

MIT