Fire Enrich is an AI-powered data enrichment tool designed for users needing to transform lists of emails into comprehensive company profiles. It leverages a multi-agent system, powered by Firecrawl and OpenAI, to gather details such as funding, tech stacks, and industry classifications, providing a richer dataset for analysis and outreach.

How It Works

The tool employs a phased, sequential multi-agent architecture. Each agent specializes in a specific data category (e.g., Discovery, Company Profile, Financial Intel, Tech Stack) and builds upon the context established by previous agents. Within each phase, multiple parallel searches are executed using the Firecrawl API to maximize efficiency and accuracy. A final GPT-4o synthesis step combines and validates findings from all agents, producing structured output with source citations. This sequential approach enhances accuracy by allowing agents to refine and validate data from earlier stages.

Quick Start & Requirements

• Install/Run: Clone the repository, create a .env.local file with FIRECRAWL_API_KEY and OPENAI_API_KEY, then run npm install (or yarn install) followed by npm run dev (or yarn dev).
• Prerequisites: OpenAI API key, Firecrawl API key.
• Demo: http://localhost:3000

Highlighted Details

• Phased Extraction System: Sequential modules build context for increasingly accurate results.
• Real-time Streaming: Data enrichment is visible row-by-row via Server-Sent Events.
• Full Source Citations: Every data point is linked to its source URL for transparency.
• Unlimited Mode: Local development enables unlimited data processing, bypassing demo restrictions.

Maintenance & Community

The project is open-source with a philosophy encouraging community contributions via pull requests and issue discussions. Support is primarily handled through GitHub issues.

Licensing & Compatibility

MIT License. This permissive license allows for commercial use and integration into closed-source projects.

Limitations & Caveats

The project is described as "just the start," implying that feature parity with mature commercial services may not be immediate. The effectiveness of enrichment is dependent on the availability and quality of data found via Firecrawl and OpenAI.