514-labs/moosestack

The agent harness for building analytics into your app on top of ClickHouse, Redpanda and other high-performance analytical infrastructure

What it solves

MooseStack is a developer agent harness designed to help AI coding agents (like Claude Code, Cursor, and Copilot) build and ship applications on real-time analytical infrastructure. It addresses the difficulty generalist agents face when dealing with specialized terrain like ClickHouse, OLAP optimization, and real-time streaming ETL/ELT pipelines.

How it works

MooseStack acts as an intermediary layer between the AI agent and the data stack. It converts infrastructure requirements into code-first declarations in TypeScript or Python, which the harness then wires up. It provides a suite of tools to make the environment AI-friendly:

  • Dev Framework & CLI: Enables rapid, safe iteration on analytical workloads.
  • Local-first Experience: Uses moose dev to create a full mirror of the production environment on a local machine.
  • Schema & Migration Management: Provides typed schemas in application code with automated migration support.
  • AI-Specific Tooling: Includes a Model Context Protocol (MCP) interface, a Language Server (LSP) for in-editor diagnostics, and a library of agent-readable rules for ClickHouse and OLAP best practices.

Who it’s for

Developers and AI coding agents who are building applications using real-time analytical infrastructure, specifically those utilizing ClickHouse, Redpanda, Temporal, and Redis.

Highlights

  • Code-First Infrastructure: Declare tables, streams, and workflows in TS/Python.
  • Modular Design: Independent modules for OLAP, Streaming, and Workflows (via Temporal) and APIs.
  • Git-native Development: Built-in version control and governance.
  • Agent-centric Tooling: Dedicated MCP and LSP to provide agents with the necessary context and context-aware autocomplete.

Related

  • Project
    am-will/gooey-pi

    A desktop workspace that provides a unified GUI for Pi, OMP, and Prime Agent coding agents, integrating coding tools, a browser, and session management.

  • Project
    sandbaseai/sandbase-harness

    A local-first runtime for AI agents that provides sandboxed code execution, persistent session management, and tool governance on the user's own infrastructure.

  • Project
    agentscope-ai/agentscope

    A production-ready agent framework that provides an SDK and a FastAPI-based service to build, deploy and scale multi-agent applications with built-in RAG and tool management.

  • Project
    deepset-ai/haystack

    An open-source AI orchestration framework for building production-ready RAG systems, autonomous agents, and LLM applications in Python.

  • Project
    i-am-bee/agentstack

    An open infrastructure for turning AI agents into running HTTP services, providing the runtime, LLM routing, and deployment tooling needed to ship agent-powered features.