datagallery-ai/dataagent

DataFoundry is an open-source AI workbench for data analysis, unifying data sources, knowledge, tools, and agent runtime into a governed workspace for interactive analytics.

What it solves

DataFoundry addresses the risks and inaccuracies associated with using AI to query enterprise databases. It moves beyond simple "prompt-to-SQL" chatbots by ensuring the AI understands business definitions (semantics), operates within strict read-only boundaries to prevent data mutation, and maintains a fully auditable trail of every step to ensure conclusions are verifiable.

How it works

The system operates as a governed data-task workbench. When a user asks a business question, the agent first aligns the natural language terms with the actual database schema and business metrics. It then executes queries through a "Data Gateway" that enforces security guardrails—such as row limits, timeouts, and field masking—before materializing the results into tables, charts, or reports. Every action, including the SQL generated and tool calls, is persisted for replay and review.

Who it’s for

It is designed for enterprise teams who need to perform complex, multi-step data analysis using AI without compromising production data security or sacrificing the accuracy of business-specific definitions.

Highlights

  • Broad Connectivity: Supports 28 datasource types out of the box, including PostgreSQL, Snowflake, BigQuery, MongoDB, and Elasticsearch.
  • Enterprise Semantics: Centralized management of schema and metric definitions to prevent the AI from guessing fields or joins.
  • () Self-Hosted & Multi-Model: Can be deployed locally to keep data within company boundaries and supports any OpenAI-compatible model (e.g., Qwen, DeepSeek, GPT).
  • Auditability: Full persistence of SQL and event streams, making every analysis step replayable and evidence-backed.
  • Flexible Interfaces: Provides a Web workbench, a Terminal UI (TUI), and a REST API for embedding the agent runtime into other products.

Related

  • Project
    darcyegb/ClaudeCodeAgents

    A collection of specialized agent definitions for Claude Code that provide focused expertise for code review, implementation verification, and UI testing.

  • Project
    Tomsawyerhu/Chinese-WebNovel-Skill

    A modular, retrieval‑augmented prompt framework for AI‑assisted Chinese web‑novel writing. It splits the workflow into a short router (`SKILL.md`) and many self‑contained sub‑skills (concept planning, opening, plot logic, dialogue, consistency review, etc.). Each sub‑skill ships tutorials, runtime rules, good/bad examples, and may use local novel corpora via a provided search script. The system routes a user’s brief idea to the appropriate module, pulls relevant example excerpts, composes a concise prompt, generates text, and finally runs a consistency review. Designed for Chinese novel authors, prompt engineers, and researchers interested in modular LLM writing pipelines.

  • Project
    ChatbotXIO/ChatbotX

    An agentic omnichannel chat marketing platform that allows users to build AI-powered chatbots with a visual flow builder and self-host their automation.

  • Project
    NVIDIA-AI-Blueprints/rag

    A reference solution for building enterprise RAG pipelines using NVIDIA NIM microservices, supporting multimodal data extraction and agentic multi-hop retrieval.