Skip to content

s3ak6i-dev/AgentRank

Repository files navigation

AgentRank

The trust and reputation layer for AI agents.

Protocols like MCP and A2A tell you how agents connect. AgentRank tells you whether you should connect.

What is this?

AgentRank is a platform where AI agent builders publish profiles for their agents, and developers/enterprises can search, compare, and review them. Think of it as the credit rating agency for the agentic economy.

Features (MVP)

  • Agent Profiles — Name, description, skills, integrations, version history, deployment stats
  • Search & Discovery — Full-text search with filters (category, skills, integration type, trust tier)
  • Reviews & Ratings — Star ratings, text reviews, helpful/not-helpful voting
  • Trust Tiers — Automated reputation tiers (bronze → silver → gold → platinum) based on review data
  • Compare — Side-by-side comparison of up to 3 agents
  • GitHub OAuth — Sign in with GitHub, pull profile data automatically

Tech Stack

Layer Technology
Framework Next.js 15 (App Router)
Language TypeScript (strict)
Database PostgreSQL (Neon)
ORM Drizzle ORM
Auth Auth.js v5 (GitHub OAuth)
Styling Tailwind CSS v4
Validation Zod
Hosting Vercel

Getting Started

Prerequisites

Setup

  1. Clone the repo:

    git clone <repo-url>
    cd agentrank
  2. Install dependencies:

    npm install
  3. Copy environment variables:

    cp .env.local.example .env.local
  4. Fill in .env.local:

    • DATABASE_URL — Your Neon connection string
    • AUTH_SECRET — Run openssl rand -base64 32
    • AUTH_GITHUB_ID / AUTH_GITHUB_SECRET — From your GitHub OAuth App
  5. Run database migrations:

    npx drizzle-kit migrate
  6. Start the dev server:

    npm run dev
  7. Open http://localhost:3000

Project Structure

src/
├── app/              # Next.js App Router pages and API routes
├── components/       # React components (ui/, agents/, reviews/, search/, layout/)
├── lib/
│   ├── db/           # Drizzle schema, relations, migrations
│   ├── validators/   # Zod input schemas
│   ├── auth.ts       # Auth.js config
│   └── utils.ts      # Shared utilities
├── hooks/            # Custom React hooks
└── types/            # Shared TypeScript types

See docs/IMPLEMENTATION_PLAN.md for the full sprint breakdown.

License

MIT

About

The trust and reputation layer for AI agents. Protocols like MCP and A2A tell you how agents connect. AgentRank tells you whether you should connect.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages