Back to directory

Composable Pipelines

A Swift package and compiler for authoring AI pipelines as a declarative, serializable graph that separates authoring, compilation and execution into distinct layers.

Composable Pipelines

Composable Pipelines is an intermediate representation and compiler for AI pipelines in Swift. It separates a pipeline into three layers: a declarative front end written with a SwiftUI-like result-builder syntax, a serializable Codable AST that acts as the intermediate representation, and a pluggable backend that executes the graph through a single Executor interface supplied by the caller.

A pipeline authored this way lowers to the AST, a compiler analyzes data dependencies between state slots to emit an execution graph, and an observable walker runs that graph with incremental, epoch-based re-execution, including native branching and looping constructs. The same primitives scale from a two-step summarization pipeline to a full tool-calling agent. Composable Pipelines is the open authoring, intermediate-representation and compiler layer of MacPaw's proprietary Elix AI engine, which supplies the production execution backend.

More in Agent frameworks

See all 902
  • Recurse

    recurse.run

    A serverless harness that lets a coding agent design, test, and deploy its own specialist sub-agents as tools, MCP servers, or bots, defined by a schema-validated manifest.

  • ProofForge

    shulgin.is-a.dev

    An AI-agent pipeline that decomposes a mathematics problem, proves each piece, and formalizes the result in Lean 4 and Mathlib until the Lean kernel accepts it, with six of its proofs already merged into Google DeepMind's formal-conjectures repository.

  • local-coder

    github.com

    Builds a local coding-agent team on top of Ollama and OpenCode by detecting the machine's hardware, recommending models for seven specialist roles, and testing whether each assigned model can actually use tools, edit a repository and pass verification.