HomeProjectsArticlesPK BoardAbout
Explore Projects

WrenAI

Active
GitHub Python NOASSERTION

Description

Open-source text-to-SQL and text-to-chart GenBI agent with a semantic layer. Ask your database questions in natural language and get accurate SQL, charts, and BI insights. Supports 12+ data sources and any LLM.

Key Features

  • Open context layer providing business semantics, examples, and governance for AI agents
  • Agent-driven design with CLI-embedded workflow guides loaded on demand
  • Modeling Definition Language (MDL) for models, relationships, cubes, metrics, and RLAC/CLAC
  • Supports 22+ data sources powered by Apache DataFusion engine
  • Agent SDK integration with LangChain/LangGraph and Pydantic AI
  • Reviewable, reproducible context with Git-friendly versioning for all definitions

Use Cases

💡 Natural language database queries generating SQL and visualizations
💡 Building business data semantic layers for AI agents
💡 Unified querying across multiple data sources for BI analysis
💡 Agent-driven data exploration and report generation
💡 Enterprise data governance and access control

Strengths & Limitations

✅ Strengths

  • • Actively maintained, recent updates
  • • High community interest (17.5k stars)
  • • Established track record (2 years in production)

⚠️ Limitations

  • • No clear open-source license

Tags

llm typescript agent data-processing api

Categories

📚 RAG Tools ⚡ Agent Tools

Quick Start

```bash
pip install wrenai
npx skills add Canner/WrenAI
```

After installing the CLI and discovery stub, tell your agent in a project directory: 'Use Wren to set up my Postgres database.' The agent runs `wren skills get onboarding` and follows the guide step-by-step. Supports the bundled `jaffle_shop` sample dataset for quick experimentation.
Visit GitHub

Related Projects

LangExtract

38.5k · Python
Active A+

A Python library by Google for extracting structured information from unstructured text using LLMs with precise source grounding and interactive visualization, designed for data annotation and knowledge extraction workflows.

data-processingllmpython +2
  • · Precise source grounding — Every extraction maps to exact source text location with visual highlighting for traceability and verification
  • · Controlled structured outputs — Few-shot example enforced output schema, leveraging Gemini controlled generation for robust results
  • · Long document optimization — Text chunking, parallel processing, and multi-pass strategy to overcome needle-in-a-haystack challenges
View full details → View on GitHub

Crawlee

25.6k · TypeScript
Active A+

A web scraping and browser automation library for Node.js to build reliable crawlers, supporting Puppeteer, Playwright, Cheerio, and raw HTTP. Extract data for AI, LLMs, RAG, or GPTs with proxy rotation and both headful and headless modes.

typescriptjavascriptdata-processing +3
  • · Multi-engine unified interface — same API supports Puppeteer, Playwright, Cheerio and raw HTTP
  • · Anti-detection stealth — default config simulates human behavior to bypass major bot protections
  • · Proxy rotation and session management — built-in proxy rotation, fingerprint generation and session persistence
View full details → View on GitHub

Docstrange

1.5k · Python
Stale B

Extract and convert data from any document (PDFs, images, Word, PPT, URLs) into multiple formats including Markdown, JSON, and CSV.

pythonragtools +2
  • · Multi-format conversion — transforms PDF, DOCX, PPTX, XLSX, images, and URLs into Markdown, JSON, CSV, and HTML
  • · 7B parameter model — upgraded core model for significantly higher accuracy and deeper document understanding
  • · Advanced OCR pipeline — extracts text from scanned documents, phone photos, and receipts with high accuracy
View on GitHub