Pipelex
Declarative language for repeatable AI workflows (MIT)
FreeOpen SourceWorkflow automationAI26,688 impressions#1 of its week2 comments
Comments
>log in to commentIf you want to use Pipelex IRL, we're hosting a Hackathon in San Francisco on Wed 10/29 Register to attend : https://luma.com/4jwfaw71
Hey DevHunt! We built Pipelex because we kept rewriting the same agentic patterns across projects. Instead of more glue code, we modeled *meaningful* steps that both humans and LLMs can read and execute like Dockerfile/SQL for AI workflows. What’s included - Python library for local dev - FastAPI server + Docker image (self-host) - MCP server (agents can *run* and even *build* pipes) - n8n node for automations - VS Code / Cursor extension (PLX syntax) What we’d love feedback on 1. Does the PLX syntax help you model your use case? 2. Agent/MCP workflows & n8n node devX. 3. Missing pipe types / model integrations. 4. OSS contributions (core + community pipes). Known limitations - Connectors: we focus on cognitive steps; bring your own app/API (or MCP/n8n). - Visualization: flow-charts WIP. - Pipe builder can fail on very complex briefs; we’re adding recursion. - No hosted API yet (on the way). - Cost tracking: LLM only for now. - Caching & reasoning options: not yet. Thanks for trying even one workflow and telling us exactly where it hurts — that’s the most valuable feedback.
Pipelex is a declarative language and Python runtime for building repeatable, agent-first AI workflows that run on any model or provider.
- for
- Developers building LLM pipelines, AI agents, and chatbot tools.
- pricing
- open source
- license
- MIT
Key features
- Declarative MTHDS language — Write typed AI methods in a Dockerfile/SQL-like syntax that defines steps, inputs, and outputs.
- Agent plugins — Claude Code and Codex plugins let agents design, run, and save methods directly from the IDE.
- Multi-environment execution — Run methods as a chatbot MCP, a webapp, or via API/CLI on any infrastructure.
- Open standard runtime — Pipelex runtime executes methods locally or in Docker, handling model routing and structured output.
- SDKs and integrations — TypeScript (@pipelex/sdk) and Python (pipelex-sdk) clients generate typed calls to the Pipelex API.
- VS Code extension — Provides syntax highlighting, linting, and formatting for .mthds files.
Use cases
- Extract structured data from PDFs via a chatbot request
- Turn an AI method into a Next.js webapp with auto-generated UI
- Integrate a reusable AI pipeline into existing Python or TypeScript code
- Run custom LLM workflows on self-hosted infrastructure for privacy
Pipelex vs alternatives
| Best for | Declarative AI workflow orchestration | TypeScript-based CI/CD pipelines | Visual backend building | AI-assisted software generation | AI coding assistance |
|---|---|---|---|---|---|
| Pricing | Open source | Free | Free | Free | Free |
| DevHunt upvotes | 40 | 158 | 118 | 75 | 63 |
| Launched | Oct 2025 | Jan 2024 | Jan 2023 | Jan 2023 | Mar 2024 |
- Pipelex vs Fluent CI: Fluent CI focuses on writing CI/CD pipelines in TypeScript, not on declarative AI workflow orchestration.
- Pipelex vs BuildShip: BuildShip provides a visual drag-and-drop backend builder, whereas Pipelex uses code-first declarative definitions.
- Pipelex vs Fine: Fine lets you build software with AI agents but lacks a dedicated declarative language and runtime for repeatable pipelines.
- Pipelex vs Pieces Copilot: Pieces Copilot offers an AI coding sidekick for workflow assistance, not a full pipeline language/runtime like Pipelex.
Pipelex FAQ
How do I start building a method?+
Sign up at app.pipelex.com, install the Pipelex plugin in Claude Code or Codex, then ask the agent to design a method using /pipelex-design.
Can I run methods on my own machine?+
Yes, the Pipelex runtime and pipelex-api Docker image let you host the runner locally and choose any model provider.
What is the MCP?+
The Pipelex MCP is a connector that lets chatbots like ChatGPT or Claude list and run saved methods directly in a conversation.
Do I need to host any services to use Pipelex?+
No, you can use the hosted Pipelex service, or self-host the runtime and API if you prefer.
What licenses apply?+
The runtime and servers are Elastic License 2.0; SDKs, extensions, and the MTHDS standard are MIT (or Apache 2.0 for some components).
Summarized by DevHunt from github.com · Sep 27, 2026. Details may change; check the official site.