---
title: "Open Source at Atlan | Projects, Upstream Work and Talks"
url: "https://atlan.com/open/"
description: "The open-source projects Atlan builds for AI agents on data, the upstream projects our engineers contribute to, and our talks and engineering writing."
---

# Open source at Atlan

Atlan runs on open source. We contribute fixes to the projects we depend on and publish the tools we build for AI agents. Atlan's open projects live at [github.com/atlanai](https://github.com/atlanai) and [github.com/atlanhq](https://github.com/atlanhq).

## Projects

Tools we built for AI agents on data, released in the open. Each one started as a problem we had running agents in production.

- [ai-registry](https://github.com/atlanai/ai-registry): an open standard for AI registries, plus the aireg CLI. Keep your team's skills and agents pinned and verified in git, then install them into Claude Code, Codex or Cursor.
- [analytics-playbook](https://github.com/atlanai/analytics-playbook): how to build an analytics agent that gets your business numbers right: metrics from your analysts' SQL, a golden set of real questions, and improvement loops.
- [decision-bench](https://github.com/atlanai/decision-bench): an open benchmark of how accurately, quickly and cheaply language models make bounded decisions on real data.
- [agent-toolkit](https://github.com/atlanhq/agent-toolkit): Atlan's open MCP server and agent toolkit, for connecting AI agents to the metadata in Atlan.

## Atlan SDKs

All four SDKs are Apache-2.0.

- [atlan-python](https://github.com/atlanhq/atlan-python), [atlan-java](https://github.com/atlanhq/atlan-java) and [atlan-go](https://github.com/atlanhq/atlan-go): SDKs for Atlan's APIs.
- [application-sdk](https://github.com/atlanhq/application-sdk): for building apps on Atlan.

## Upstream

The projects Atlan runs on, and contributes back to.

- [DuckDB](https://github.com/duckdb/duckdb): the in-process analytical database. Atlan engineers contribute to DuckDB core, duckdb-iceberg and duckdb-rs.
- [OpenLineage](https://github.com/OpenLineage/OpenLineage): the open standard for collecting lineage from data pipelines and orchestrators.
- [Apache Ossie](https://github.com/apache/ossie): the open format for semantic models: datasets, relationships, metrics and context for agents. Atlan's first contributions are in review.

## Other projects our engineers send fixes and features to

### Data and pipelines

- [Apache Airflow](https://github.com/apache/airflow): workflow orchestration.
- [Argo Workflows](https://github.com/argoproj/argo-workflows): workflow engine for Kubernetes.
- [Daft](https://github.com/Eventual-Inc/Daft): distributed query engine.
- [SQLGlot](https://github.com/tobymao/sqlglot): SQL parser and transpiler.

### AI and agents

- [ZeroClaw](https://github.com/zeroclaw-labs/zeroclaw): agent runtime.
- [vLLM Semantic Router](https://github.com/vllm-project/semantic-router): model routing for vLLM.
- [Google ADK docs](https://github.com/google/adk-docs): Agent Development Kit documentation.
- [Laravel AI](https://github.com/laravel/ai): AI toolkit for Laravel.
- [Screenpipe](https://github.com/screenpipe/screenpipe): screen and audio capture for AI.

### Infrastructure and tooling

- [KEDA](https://github.com/kedacore/keda): event-driven autoscaling for Kubernetes.
- [Dapr](https://github.com/dapr/dapr): distributed application runtime.
- [Terraform AWS provider](https://github.com/hashicorp/terraform-provider-aws): AWS resources in Terraform.
- [VictoriaMetrics operator](https://github.com/VictoriaMetrics/operator): VictoriaMetrics on Kubernetes.
- [Tauri plugins](https://github.com/tauri-apps/plugins-workspace): official Tauri plugins.
- [Markdown Guide](https://github.com/mattcone/markdown-guide): Markdown reference.

## Talks and writing

What our engineers learned building in the open.

- 30 Oct 2026, article: [Introducing ai-registry](https://blog.atlan.com/engineering/introducing-ai-registry/), Atlan engineering blog.
- 28 Oct 2026, talk: [Open sourcing a spec for AI registries](https://atlan.com/context-conference/lightning-talk/), Atlan Context Conference.
- 23 Sep 2026, article: [Zero Graph: removing the graph database](https://blog.atlan.com/engineering/zero-graph-removing-the-graph-database/), Atlan engineering blog.
- 23 Sep 2026, article: [Empathy at Scale: an LLM judge for MCP](https://blog.atlan.com/product/llm-judge-mcp-server/), Atlan product blog.
- 10 Aug 2026, article: [Measuring user value on an MCP server](https://blog.atlan.com/product/measuring-user-value-on-an-mcp-server/), Atlan product blog.

## Get involved

Send a pull request, or tell us what you're building. Start at [github.com/atlanai](https://github.com/atlanai) or read the [ai-registry contributing guide](https://github.com/atlanai/ai-registry/blob/main/CONTRIBUTING.md).

## Next steps

- To see Atlan's context layer in action first, watch a recorded demo: [Atlan context layer demos](https://atlan.com/context-layer-demo/)
- To see Atlan working on your own data stack, book a demo: [Book a demo with Atlan](https://atlan.com/forms/talk-to-sales-contact/)
