Pipelex: Declarative Language for Composable AI Workflows
tom_doerr · x · 2026-08-11
Pipelex is an open-source project that provides a declarative language for building composable AI workflows, allowing AI methods to be run as executable pipelines. The project has 696 stars on GitHub, 1,263 commits, and offers documentation, a demo, and a hub. Its core concept is the 'method', a reusable, typed AI procedure declared in .mthds files and executed by Pipelex, with explicit steps, structured outputs, and repeatable runs.
More from coding & agent
- Lindy Teammate Integrates Directly Into Slack with Persistent Memory — SucceededMind · 2026-08-11
- Open-mercato: Open-Source Framework to Align AI Coding Agents with Project Architecture — tom_doerr · 2026-08-11
- shadcn Open Sources Minimal AI Chatbot Template with Tool Calling & HITL — shadcn · 2026-08-11
- Revision Prompting: Cut LLM Processing Costs by 65% Using Input Diffs — Dry_Rabbit_1123 · 2026-08-11
- Asana Bets on AI Agents with $75M Acquisition of Stack AI — shashib · 2026-08-11
- Paper: Do Frontier Models Eliminate the Need for Prescribed Pipelines in Self-Evolving Agents? — burny_tech · 2026-08-11