Data Connector – Chat with Your Database and APIs
Hey HN! Wanted to share a project I have been working. Data Connector is an open source tool for interacting with Databases (Postgres, MySQL, SQLite) using natural language. Prompts are handled using a ReAct[1] agent which has access to the schema and a tool for executing queries. It also has experimental[2] support for calling APIs (OpenAPI, GraphQL) which allows for queries that can span multiple systems (Find this user's details in Clerk, etc). We have mostly been using it for managing our local development environment, but it has configurable controls for running in other environments. -…
In plain words
Data Connector is an open-source tool that lets users query databases and APIs using natural language. It works with Postgres, MySQL, SQLite, and supports experimental API integration through OpenAPI and GraphQL, enabling queries across multiple systems. Built with a ReAct agent that accesses database schemas and executes queries, it includes privacy mode to keep results from the language model and approval mode for manual query verification. Designed primarily for development environments with configurable controls for other uses.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
Hey HN! Wanted to share a project I have been working. Data Connector is an open source tool for interacting with Databases (Postgres, MySQL, SQLite) using natural language. Prompts are handled using a ReAct[1] agent which has access to the schema and a tool for executing queries. It also has experimental[2] support for calling APIs (OpenAPI, GraphQL) which allows for queries that can span multiple systems (Find this user's details in Clerk, etc). We have mostly been using it for managing our local development environment, but it has configurable controls for running in other environments. - Privacy Mode which never sends query results back to the LLM context, instead the result is returned to the user as JSON - Approval Mode which requires manual approval of each query before it is executed Data Connector runs within a private subnet, connects to the database and polls for new queries. There is no requirement to open traffic to the database or share credentials / connection strings. Data Connector is built using another project we are working on Inferable (also open source) which is a runtime for building and observing LLM-based agents. Would love your thoughts! [1] https://arxiv.org/abs/2210.03629 [2] We have been testing these with various schemas and are currently working on better handling for large OpenAPI specs, etc.
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 26d ago · cactuscompute.com
