📡 MCP Radar

LeastAction-Labs/LeastAction-Lite

One platform to build, run, and operate every data workflow your team touches. Source-available, self-hosted.

4 stars
0 forks
Python
momentum ▲ 6.7
created 2026-08-21
on radar since 2026-08-25
aiautomationcatalogdata-engineeringdata-flowdata-integrationdata-pipelinesdevelopmentetlintegration-frameworkintegrationsleastactionlow-codelow-code-platformmcpmcp-clientmcp-serverself-hostedworkflowworkflow-automation
View on GitHub ↗Homepage ↗

About LeastAction-Lite

Most AI tools help you write pipeline code. LeastAction gives the AI an operator — it generates the code, deploys it, runs it, reads the logs, queries the data, spots the issue, fixes the operator, and reruns. No human in the loop at each step. It works against whatever stack you already have — PostgreSQL, Athena, Redshift, BigQuery, S3, Lambda, dbt, Airflow, any API — with no migration.

Everything lives in a catalog: operators, connections, payloads, configs, actions, tasks, reports, and assets. The catalog is the shared context the AI works from — and the same interface engineers use to schedule, monitor, and reuse pipelines across teams. Self-hosted, one instance per team.

Airflow, Dagster, and Prefect are excellent at running pipelines. LeastAction takes a different starting point — an AI operator and a catalog open to the whole team:

- The AI operates, end to end. It generates the operator, deploys it, runs it, reads logs, queries the result, fixes its own bug, and reruns — not just code suggestions, but the full loop. - Orchestrate without Python. Engineers write operators in Python (or generate them with AI); everyone else assembles, schedules, and runs pipelines from the UI or Git. - Custom operators and connections, no package infrastructure. Write an operator, save it to the catalog, use it immediately — no provider packages, no per-worker deploy. Connections take any fields your operator needs. - Granular control built in. N-level config with locked/overridable parameters, per-connection parallelism with priority queuing, native Git CI/CD, and 1-click backfill

From the project README.

Maintaining this server?

Add the radar badge to your README — it shows your project was picked up by MCP Radar and links to this page:

[![On MCP Radar](https://mcp.liqiwa.com/badge.svg)](https://mcp.liqiwa.com/s/LeastAction-Labs--LeastAction-Lite.html)

Related MCP servers

iflytek/dolphin-mcp-pilot

Production-ready MCP server that lets AI agents build, schedule, run and recover Apache DolphinScheduler data pipelines - 53+ tools, beyond read-only.

⭐ 7Python
anhnv02/file-mcp

FileMCP is a native macOS app that gives ChatGPT controlled access to a local workspace through MCP

⭐ 6Swift
Dragkob/MCPhantom

MCPhantom is a local web dashboard for MCP endpoint recon. Discover resources, templates, tools, and prompts, then interact with them from a security console.

⭐ 6Python
herry2059/project-os-for-codex

Open-source control plane for Codex projects: Git-backed context, visible agent progress, scoped MCP access, resumable work, and safe handoffs.

⭐ 104TypeScript
desktop-commander/remote-desktop-commander

Official remote MCP server for Desktop Commander — connect claude.ai, ChatGPT, Cursor and more to your own computer over OAuth

⭐ 3
KB-perByte/hiveshare

Collaborative AI memory for engineering teams - share context across Claude Code, Cursor, and teammates

⭐ 3Go

📬 Get the weekly radar in your inbox

The top new MCP servers of the week, every Monday. No spam, unsubscribe anytime.