DEDUPED Hot AI repos. Real use cases. What people build. Get the Monday 5

How to turn your company's databases and docs into one queryable graph with Helix Foundry

Run Foundry locally, connect a database, let the agent draft an ontology and load it into HelixDB.

Short answer: Run git clone https://github.com/helixdb/helix-foundry.git and cd helix-foundry; then start it with docker compose up -d --wait (export ANTHROPIC_API_KEY or OPENAI_API_KEY first if you want Claude or OpenAI); open http://localhost:3000; there is no sign-in. All steps.

data-access. Foundry reads the databases you connect; use a read-only database user and the sample data first.

Who it is forAn ops or data lead whose customers, invoices and tickets live in different systems.

The problemAnswering cross-system questions means manual joins and nobody trusts where the numbers came from.

The outcomeA HelixDB graph you can explore, with answers that cite the row or quote behind each value.

What you need first

  • Docker with Compose v2 to run locally
  • Optional: ANTHROPIC_API_KEY or OPENAI_API_KEY in the shell for Claude or OpenAI
  • Node 24 and pnpm only for developing the site

Steps

  1. Run git clone https://github.com/helixdb/helix-foundry.git and cd helix-foundry
  2. Start it with docker compose up -d --wait (export ANTHROPIC_API_KEY or OPENAI_API_KEY first if you want Claude or OpenAI)
  3. Open http://localhost:3000; there is no sign-in
  4. Connect PostgreSQL or another source, or choose Use the sample for the Acme Corp demo
  5. Review and edit the ontology the agent drafts, make a version current, and load it

Prompt for your agent

Run HelixDB/helix-foundry locally with Docker Compose, connect <database>, review the ontology the agent drafts, load it, and ask: <question>. Show the cited rows behind the answer.

Caveat. This card is written from the HelixDB/helix-foundry README, not from a test run by Deduped. Before you start, make sure you have Docker with Compose v2 to run locally. Check the repo for newer instructions before you rely on these steps.

Use this in your AI agent

Ask your agent

Using the Deduped MCP, get the setup steps for HelixDB/helix-foundry and "Turn your company's databases and docs into one queryable graph"

Add Deduped to Claude Code once

claude mcp add --transport http deduped https://dedupedai.com/mcp

Cursor, ChatGPT, others

Share: X LinkedIn Hacker News Email Copy link

Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer