Skip to main content
Glama
skyzhao1223

skyzhao1223/wheel-hub

⚙️ wheel-hub

Stop reinventing the wheel. Search, score and adopt the best open-source solutions — before you write a line of code.

wheel-hub is an MCP server + CLI that makes both AI coding agents and humans check the open-source ecosystem first: given a feature or problem, it finds existing solutions and scores them on popularity, momentum, maintenance and trust, then tells you whether to adopt, adapt, inspect or avoid.

Why

Every developer (and every AI agent) rebuilds auth, schedulers, scrapers and admin panels from scratch — not because better wheels don't exist, but because nobody looks for them first. wheel-hub makes "look first" the default.

Related MCP server: reuse-before-generate

Install

MCP server (for Claude Code, opencode, Cursor, …)

npx @wheel-hub/mcp

Claude Code example:

claude mcp add wheel-hub --env GITHUB_TOKEN=ghp_xxx -- npx @wheel-hub/mcp

Generic JSON config example (works in most MCP clients):

{
  "mcpServers": {
    "wheel-hub": {
      "command": "npx",
      "args": ["@wheel-hub/mcp"],
      "env": { "GITHUB_TOKEN": "ghp_xxx" }
    }
  }
}

Note: MCP clients only forward a whitelist of environment variables to servers. Always pass GITHUB_TOKEN explicitly through the client's env config, or the server will run unauthenticated at 60 requests/hour and hit rate limits quickly.

Then just ask your agent: "before you implement, use wheel-hub to check what exists for background job scheduling".

CLI (for humans)

# one-off
npx @wheel-hub/cli find "SSO login" --lang typescript
npx @wheel-hub/cli evaluate ollama/ollama

# or install for the `wheel-hub` command
npm i -g @wheel-hub/cli
wheel-hub find "SSO login"

Tools

Tool

Description

search_oss

Search GitHub for solutions to a problem

evaluate_repo

Score one repo (0–100) with a verdict and reasons

compare_alternatives

Find alternatives + evaluate each — run this before writing new code

Scoring model

Composite score (0–100) weighted across four dimensions:

  • Popularity (30%) — star mass on a log scale

  • Momentum (25%) — stars/day growth + recency of pushes

  • Maintenance (30%) — commit freshness + issue backlog ratio

  • Trust (15%) — license permissiveness, archived status

Verdicts: ≥75 adopt · ≥55 adapt · ≥35 inspect · <35 / archived avoid

Configuration

GITHUB_TOKEN (optional) — raises the GitHub API rate limit from 60 to 5,000 requests/hour. Strongly recommended for real use; see the MCP config above for how to pass it.

Packages

Package

Description

@wheel-hub/core

GitHub aggregation + scoring engine

@wheel-hub/mcp

MCP server for AI agents

wheel-hub

Terminal CLI

Roadmap

  • License fallback for NOASSERTION repos: parse the license file itself (e.g. the kernel's COPYING declares GPL-2.0 but GitHub classifies it as "Other")

  • Awesome-list mining as a secondary signal source

  • Star-history curves via OSSInsight data

  • wheel-hub watch — get notified when your stack's dependencies go stale

  • Score persistence + weekly digest

License

MIT

Related MCP Connectors

Related MCP Servers

  • A
    license
    Not graded
    quality
    B
    maintenance
    A local MCP server that gives AI agents access to developer tooling — GitHub (read-only), documentation search, and web research — via stdio transport.
    MIT
  • A
    license
    Not graded
    quality
    A
    maintenance
    MCP server that helps coding agents research and prepare GitHub contributions by gathering repository guidance, related work, code context, and validation evidence into a local SQLite corpus, without ever writing to GitHub.
    246
    3
    MIT

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/skyzhao1223/wheel-hub'

If you have feedback or need assistance with the MCP directory API, please join our Discord server