yuki-20

yuki-20 / CornMCP

Public

CornMCP is an open-source mono repo that gives AI coding agents (Antigravity, Cursor, Claude Code, Codex), token-efficient access to your codebase through the Model Context Protocol (MCP).

18
1
100% credibility
Found Mar 29, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

CornMCP is a self-hosted platform that equips AI coding agents with precise codebase understanding, cross-session memory, enforced quality checks, and live performance analytics through a simple local setup.

How It Works

1
🔍 Discover Corn Hub

You hear about a helpful tool that supercharges your AI coding buddy by giving it deep knowledge of your entire project.

2
📥 Bring it home

Grab the tool and get it running on your computer with a few easy steps, everything stays local and private.

3
🔗 Link to your coding app

Connect it to your favorite AI coding helper so it can tap into smart features like code maps and memory.

4
📁 Share your project

Tell the tool about your code folder, and it starts learning the structure and connections inside.

5
🧠 Watch it build smarts

It creates a clever map of your functions, classes, and how they work together, ready for your AI to use.

🚀 AI coding unlocked

Your AI now searches code precisely, remembers past work, checks quality automatically, and saves you hours of fumbling.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 18 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is CornMCP?

CornMCP is a TypeScript mono repo delivering a local MCP server and analytics dashboard that gives AI coding agents like Antigravity, Cursor, Claude Code, and Codex token-efficient access to your codebase. It solves bloated context windows by providing 18 tools for semantic memory, AST-powered code intelligence, quality gates, and session tracking—all via Model Context Protocol over HTTP. Run it with pnpm and SQLite for surgical codebase queries without external services.

Why is it gaining traction?

It stands out by cutting token burn 30-87% through precise symbol search, call graphs, and impact analysis, letting agents grasp architecture without full-file reads. Real-time dashboards track tool latency and savings, while multi-agent awareness flags peer changes instantly. Quality enforcement blocks low-score plans, making agent outputs reliable for production coding.

Who should use this?

Cursor and Claude Code users wrestling token limits on medium/large repos. Teams running Antigravity or Codex agents collaboratively, needing change detection and shared knowledge to avoid conflicts. Devs evaluating MCP for local agent setups with built-in analytics.

Verdict

Solid MCP starter for agentic coding workflows—try it if you're in Cursor/Claude daily. Low 1.0% credibility from 18 stars signals early maturity; docs shine but test edge cases before prod.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.