Sourc

An open, self-hosted dashboard for AI-assisted startup sourcing and deal-flow management.

Bring your own model. Bring your own data. Run it locally in two minutes.

---

What it is

Sourc is a CRM-style pipeline for sourcing early-stage startups: import companies from multiple sources, score them across structured dimensions, track outreach and calls, and generate presentation-ready briefs and outreach emails with the LLM of your choice.

It runs entirely on your machine against a local SQLite database — no accounts, no SaaS, no vendor lock-in. The AI layer is provider-agnostic: plug in OpenAI, Groq, OpenRouter, DeepSeek, a local Ollama model, or anything else that speaks the OpenAI chat-completions API.

Features

• Pipeline — Kanban-style deal flow across 11 stages (New → Selected / Hell No / Already Sourced).
• Startup CRM — full records with 10 scoring dimensions, founders, outreach events, and call notes.
• AI briefs — turn a startup record into a polished, VC-standard sourcing brief (streamed).
• AI outreach — draft and refine founder emails in a consistent voice (streamed).
• Imports — load startups from accelerator lists, the Italian Registro Startup Innovative, Firecrawl scrapes, or a custom JSON queue.

Tech stack

Layer	Choice
Framework	Next.js 16 (App Router) + React 19
Database	SQLite via better-sqlite3 + Drizzle ORM
AI	Vercel AI SDK v6 — any OpenAI-compatible provider
UI	Tailwind v4 + shadcn/ui

Quick start

git clone <your-fork-url> sourc && cd sourc
npm install
cp .env.example .env.local      # then fill in your AI provider (see below)
npm run db:migrate              # creates a fresh, empty data/sourcing.db
npm run dev

Open http://localhost:3000. The database starts empty — add startups manually or via the import tools on the /sources page.

Configuring the AI provider

Sourc talks to any OpenAI-compatible chat-completions endpoint. Set three variables in .env.local — that's the only AI configuration:

AI_BASE_URL=https://api.openai.com/v1   # your provider's base URL
AI_API_KEY=sk-...                        # your key
AI_MODEL=gpt-4o-mini                     # any model the provider serves

Common providers (see .env.example for the full list):

Provider	AI_BASE_URL	Example AI_MODEL
OpenAI	https://api.openai.com/v1	gpt-4o-mini
OpenRouter	https://openrouter.ai/api/v1	openai/gpt-4o
Groq	https://api.groq.com/openai/v1	llama-3.3-70b-versatile
DeepSeek	https://api.deepseek.com/v1	deepseek-chat
Ollama	http://localhost:11434/v1	llama3.1 (any key value)

If these aren't set, the app still runs — the AI brief/email endpoints just return a clear "AI not configured" message until you add them.

Database

npm run db:generate   # generate a migration after editing lib/db/schema.ts
npm run db:migrate    # apply migrations
npm run db:studio     # open Drizzle Studio to browse data

The database file (data/*.db) is git-ignored — your data never leaves your machine and is never committed.

Project structure

app/            Next.js routes — pages + API endpoints (briefs, emails, imports)
lib/ai.ts       Provider-agnostic model resolver (the only AI config point)
lib/db/         Drizzle schema, migrations, and queries
lib/actions/    Server actions (startups, outreach, briefs, sources, ...)
components/     UI (shadcn/ui + Tailwind)