Inconvo – Chat with Your Database Without Writing SQL

TL;DR

TL;DR: Inconvo is an open-source platform for building data agents that answer natural-language questions over production databases — with safe query validation, row-level permissions, and structured outputs your application can consume directly.

What Is Inconvo?

Inconvo is an open-source platform (Apache 2.0) for building data agents — services that let your application answer natural-language questions over live production data. Instead of writing SQL or building a rigid query interface, you call an Inconvo agent and get a structured response.

A data agent in this context is not a general chatbot. It is a purpose-built service that sits between your application and your database, enforcing permissions and query constraints at every turn.

The project was founded by the Inconvo team (Y Combinator S23) and the source is available on GitHub at github.com/inconvoai/inconvo. The current release is v2.4.4.

Key Features

Safe queries — All generated queries are validated and constrained to explicitly allowed tables, columns, and joins before execution. The agent cannot query anything outside the defined schema boundaries.

Permissions and multi-tenancy — Row-, table-, and column-level access is enforced automatically. Tenant context applies at runtime without custom query logic.

Stateful interactions — Agents retain filters and refinements across conversation turns without manual state management on your end.

Observability — Every run is traceable. You can inspect the generated queries, execution logs, and failures after the fact.

Semantic modeling — Start querying immediately, then layer in business context — metrics, terminology, computed fields, join rules — over time.

Setup Workflow

Step 1: Try Inconvo Cloud

The fastest path is signing up at app.inconvo.ai and following the onboarding guide.

Step 2: Run Locally

If you prefer self-hosting for evaluation, run:

npx inconvo@latest dev

Open the dashboard at http://localhost:26686.

Step 3: Integrate into Your Application

import "dotenv/config";
import Inconvo from "@inconvoai/node";

const inconvo = new Inconvo({
  apiKey: process.env.INCONVO_API_KEY,
});

const agentConvo = await inconvo.agents.conversations.create("agt_123", {
  userIdentifier: "user_123",
  userContext: {
    organisationId: 1,
  },
});

const agentResponse = await inconvo.agents.conversations.response.create(
  agentConvo.id,
  {
    message: "What is our best selling product this week?",
    stream: false,
  },
);

console.log(agentResponse);
// { "type": "text", "message": "Your most popular product is iPhone 15." }

Source and Accuracy Notes

• Project page: inconvo.com
• Source repository: github.com/inconvoai/inconvo
• License: Apache 2.0 (verified via LICENSE file on main branch)
• Latest release: v2.4.4 (verified via GitHub Releases API, published 2026-04-07)
• HN launch thread: news.ycombinator.com/item?id=47078279
• Source last checked: 2026-06-23

FAQ

Q: Does Inconvo generate raw SQL and execute it? A: No. Inconvo explicitly avoids generating raw SQL. Instead, it validates and constrains all queries to allowed tables, columns, and joins before execution — making it safer for customer-facing applications where SQL injection or unauthorized access is a concern.

Q: What databases does Inconvo support? A: Inconvo is designed to work with standard relational databases. The semantic modeling layer lets you define how tables and columns map to business concepts, making it adaptable to different schemas.

Q: Is there a self-hosted option? A: Yes. You can run Inconvo locally using npx inconvo@latest dev. For production self-hosting, the cloud offering handles infrastructure, or you can explore the open-source codebase for self-deployment options.

Q: How does multi-tenancy work? A: Tenant context is applied at runtime. The agent enforces row-, table-, and column-level access automatically based on the userContext passed in the API call — without requiring custom query logic per tenant.

Q: What is the pricing model? A: Inconvo Cloud offers a free tier for getting started. See app.inconvo.ai for current pricing details.

Conclusion

Inconvo solves a specific problem: giving non-technical users a safe, natural-language interface to production data without exposing raw SQL or losing control over what gets queried. Its permission model and query validation are the standout features for developer teams building customer-facing data products.

If you are evaluating data agent platforms for an internal tool or customer-facing product, Inconvo is worth a look — especially if query safety and multi-tenancy are hard requirements.