ToolExperimental

datafoundry

Name: datafoundry
Rating: 3.1 (704 reviews)
Author: datagallery-lab

by datagallery-lab

Interactive AI workbench that runs and inspects data-analysis agents

TypeScript

Updated Aug 6, 2026

704

Stars

Forks

View on GitHub

What It Does

Provides an open-source AI workbench that unifies data sources, knowledge, tooling, and an agent runtime for interactive analytics. Combines a browser-based workspace with agent execution so users can run, inspect, and iterate on data-analysis agents alongside the data they act on. Distinctive features include a governed workspace model and built-in connectors for data sources and analytic tooling. This design aligns with patterns like the Orchestrator-Worker Pattern and the Event-Driven Agent Pattern to support scalable agent orchestration.

Key Benefits

As agents take on more data-heavy tasks, having a governed workspace where runs, data access, and decisions are visible is essential for trust. DataFoundry makes agent actions reproducible and auditable in the context of the datasets they touch, which helps form an agent track record and supports accountability. This kind of workbench is a practical step toward continuous agent evaluation and safer agent delegation in analytics workflows. This context resonates with the Emergence-Aware Monitoring Pattern, which emphasizes proactive visibility and safeguards.

Best For

Teams building and iterating data-analysis agents who need a governed, inspectable workspace for reproducible runs and dataset-aware debugging. For guidance on structuring such workflows, the Planning Pattern provides a useful blueprint.