All 82 Dependencies

Every package WrenAI depends on, ranked by repo health score.

WrenAI (built by Canner) is an open-source GenBI (Generative BI) engine that gives AI agents a governed path from a plain-English question to trustworthy SQL, a chart, and a deployable dashboard. Rather than letting an LLM guess at your schema, WrenAI sits underneath the agent as a context layer: a version-controlled semantic model (MDL) that captures business definitions, approved joins, units, and examples, so every agent that touches the database inherits the same vetted understanding instead of re-learning it from scratch in a prompt.

The project is delivered as an agent-driven CLI (`pip install wrenai`) rather than a hosted app. Agents such as Claude Code, Cursor, or any MCP-capable client install a lightweight discovery stub (`npx skills add Canner/WrenAI`) that teaches them to pull workflow guides on demand — onboarding a warehouse, enriching business context, generating MDL, or building a GenBI app — so the guidance always matches the installed CLI version instead of drifting out of sync with static docs.

Under the hood, SQL planning and execution run through `wren-core`, a Rust engine built on Apache DataFusion that validates and rewrites queries against the MDL before they ever touch a warehouse (dry-plan validation, structured error codes, CTE rewriting). A WebAssembly build of that same engine, `wren-core-wasm`, lets deployed dashboards query static Parquet/DuckDB snapshots entirely client-side, so a generated GenBI app can be shipped to Vercel or Cloudflare Pages with zero backend.

WrenAI connects to 20+ data sources out of the box — BigQuery, Snowflake, PostgreSQL, MySQL, ClickHouse, Redshift, Databricks, Trino, Spark, and more — through a pluggable connector layer, and ships an MCP server so any MCP-aware agent can query through the same governed semantic layer instead of hitting the raw database directly.

14 pyarrow-hotfix >=0.6 31 console_error_panic_hook 0.1 43 bigdecimal-rs 0.4.3 45 num_cpus 1.16.0 45 structopt 0.3.26 50 half 2.4.1 52 sqllogictest 0.28.4 outdated 53 Jinja >=3.1 54 csv 1.3.0 56 PyYAML >=6.0 58 pyasn1 >=0.6.3 62 rich 66 python-dotenv >=1.0 66 python-multipart >=0.0.31 67 url 2 68 base64 0.22.1 outdated 68 parking_lot 0.12.3 69 itertools 0.14.0 outdated 70 bytes 1 74 petgraph 0.8.2 75 pyOpenSSL >=26.0.0 75 tempfile 3.10.1 76 loguru >=0.7 76 rust_decimal 1.27.0 76 serde_json 1.0.117 76 thiserror 2.0.3 77 async-trait 0.1.80 77 Chrono 0.4 77 log 0.4.22 77 quote 1.0 77 serde 1.0.201 78 env_logger 0.11.5 80 SQLGlot >=29 81 duckdb >=1.5.0 81 Psycopg >=3 81 pytest-mock >=3 81 regex 1.10.5 83 object_store 0.13.1 outdated 83 Requests >=2.33.0 83 sqlparser 0.58.0 outdated 84 cryptography >=48.0.1 84 Apache DataFusion 53 outdated 84 Apache DataFusion 53.0.0 outdated 85 python-oracledb >=2 86 PyAthena >=3 86 serde_with 3.11.0 87 PySpark >=3.5 88 Pygments >=2.20.0 89 Databricks SQL Connector for Python 89 lxml >=6.1.0 90 boto3 >=1.26 90 Clap 4.4.8 90 python-sdk >=1.19 90 Pydantic AI >=1.0,<2.0 outdated 90 Ruff >=0.4 90 Typer >=0.12 90 Uvicorn >=0.29 91 Databricks SDK for Python 91 PyO3 0.29.0 92 clickhouse-connect >=0.8 92 futures 0.3.17 92 LangChain >=1.0 92 LangChain DeepSeek >=0.3 92 LangGraph >=1.0 92 Pydantic >=2 92 sentence-transformers >=3.0.0 92 snowflake-connector-python >=3.10 92 syn 2.0 outdated 92 urllib3 >=2.7.0 93 wasm-bindgen 0.3 outdated 93 Tokio 1.46.0 93 wasm-bindgen 0.2 93 wasm-bindgen 0.4 outdated 93 wasm-bindgen 0.3 outdated 94 Starlette >=1.3.1 95 Arrow 58.1 outdated 95 Arrow 58.1 outdated 95 PyArrow >=14 96 pytest >=8 96 TypeScript ^5.0.0 outdated 97 google-auth 98 pandas >=2

All 82 Dependencies

Join founders buildingwith open source

Search