The Civic Tech Field Guide offers a secure, high-performance API for developers and researchers to integrate our dataset of over 16,000 civic tech projects directly into AI agents and analytics pipelines using the Model Context Protocol (MCP).

By connecting your MCP server to our endpoints, your LLMs (like Google Gemini, Claude, or OpenAI) can directly query, embed, and perform semantic searches across our entire verified database in real-time. This powers use-cases like Retrieval-Augmented Generation (RAG), automated data analysis, trend discovery, and intelligent querying.

How It Works

Rather than relying on outdated web scraping or inefficient keyword searches, our system provides a standardized, machine-readable pipeline:

• Bulk Export API: A memory-efficient streaming endpoint (https://ctfg-api-worker.devin-a8e.workers.dev/api/v1/projects/export) that provides instant access to all ~16,000+ projects in a single, paginated JSON schema.
• Standardized Schema: All projects are normalized with title, description, url, repository_url, and flat arrays for categories and tags.
• Bring Your Own Vector DB: Ingest our raw data locally, generate your own embeddings (e.g. via Gemini text-embedding-004), and perform lightning-fast semantic similarity searches entirely within your secure environment.
• AI-Native: Provide the exact retrieved JSON structures to your LLM context window for highly accurate, hallucination-free responses.

Getting Started

The public read endpoints (export and search) require no API key — you can start querying them right away. Only sync and admin endpoints require authentication.

If your use case needs sync or admin access, please contact us at info@civictech.guide with a brief description of your project or agent.