AgentStore

The App Store for Agents.

AgentStore is a student-built marketplace where AI agents can be discovered, built, shared, purchased, installed, remixed, and improved by both humans and other agents.

Cohorts

This project is designed for two TechX cohorts:

Cohort	Focus Areas
LLM Cohort	Agent design, prompt templates, MCP/tool-calling workflows, agent manifests, tool traces, tool ecosystems, agent execution flow
Data Science Cohort	Agent ranking, recommendation systems, ratings, review sentiment, usage analytics, quality scoring, trending algorithms, marketplace insights

Architecture

Frontend (React + Vite)
        ↓
Backend API (FastAPI)
        ↓
Agent Registry → Tool Registry → Agent Store Catalog
        ↓
Agent Runner / MCP Simulation
        ↓
Tool-Call Trace Logs → Ratings, Reviews, Analytics

See docs/architecture.md for details.

Getting Started

Prerequisites

• Node.js 18+
• Python 3.11+
• Git

Backend (placeholder)

cd backend
python -m venv venv
source venv/bin/activate   # Windows: venv\Scripts\activate
pip install -r requirements.txt
uvicorn app.main:app --reload

API docs: http://127.0.0.1:8000/docs

Frontend (placeholder)

cd frontend
npm install
npm run dev

App: http://localhost:5173

Data Science

cd data-science
pip install -r requirements.txt
# TODO: Run analytics scripts or open notebooks

Repository Structure

├── docs/           # Product docs, architecture, user stories
├── frontend/       # React + TypeScript + Vite UI
├── backend/        # FastAPI REST API
├── agents/         # Agent manifests, registry, runner, traces
├── tools/          # Tool manifests and registry
├── data-science/   # Analytics, ranking, recommendation scripts
├── datasets/       # Mock CSV/JSON datasets
├── shared/         # Shared schemas and types
├── tests/          # Pytest and frontend test placeholders
├── scripts/        # Utility scripts
├── kanban/         # Task board references
└── presentations/  # Demo and pitch materials

Contributing

Students implement features through assigned Kanban tickets using branches and pull requests.

Branch Naming

feature/<ticket-id>-short-description
fix/<ticket-id>-short-description
docs/<ticket-id>-short-description

Examples:

• feature/AS-12-agent-detail-page
• fix/AS-34-rating-endpoint
• docs/AS-01-update-architecture

Pull Request Expectations

1. One ticket per PR when possible
2. Include a clear description of what changed and why
3. Reference the Kanban ticket ID
4. Keep changes focused — no unrelated refactors
5. Update relevant README or docs if behavior changes
6. Add or update tests when implementing logic

See CONTRIBUTING.md and STUDENT_ONBOARDING.md for more.

Documentation

• PROJECT_OVERVIEW.md — Product vision and concepts
• STUDENT_ONBOARDING.md — Getting started as a student developer
• ROADMAP.md — Planned milestones
• docs/product-requirements.md — MVP scope
• docs/user-stories.md — User stories
• docs/architecture.md — System architecture

License

MIT — see LICENSE.