pipeline

Orchestrate multi-stage OMX workflows with persistent state and resume capability.

Updated Apr 3, 2026

Search Tags:#api #resume #pipeline #stages #orchestrator #state-persistence #omx

Ready to run

npx skills add https://github.com/pextacy/stellar --skill pipeline-pextacy

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: pipeline
Source: https://github.com/pextacy/stellar/tree/main/skills/pipeline
Command: npx skills add https://github.com/pextacy/stellar --skill pipeline-pextacy

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The pipeline skill solves the challenge of coordinating multi-stage OMX workflows by providing a uniform stage interface and built-in persistence to resume interrupted runs.

Core Features & Use Cases

Uniform PipelineStage interface with run(ctx) and optional canSkip for intelligent stage skipping.
Built-in stages: ralplan, team-exec, and ralph-verify, and a default autopilot pipeline sequence.
State management and resume capabilities via on-disk state at .omx/state/pipeline-state.json, enabling recovery after failures.
API surface including createAutopilotPipelineConfig and runPipeline for programmatic pipeline construction and execution.
Real-world use: quickly assemble end-to-end OMX workflows for planning, execution, and verification with minimal boilerplate.

Quick Start

Define your pipeline with the API and run it to execute the configured stages.

Frequently Asked Questions about pipeline

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I configure and run an OMX pipeline with multiple stages?▼

Configure an OMX pipeline by defining stages through the createAutopilotPipelineConfig API and execute them using the runPipeline function, assembling end-to-end workflows for planning, execution, and verification with minimal boilerplate.

Can I resume an interrupted multi-stage workflow after a failure?▼

Resume interrupted multi-stage workflows using on-disk state persistence at .omx/state/pipeline-state.json, which enables reliable recovery and allows pipeline runs to continue from the last successful stage after failures.

How does stage skipping work in automated pipeline orchestration?▼

Automated pipeline orchestration handles stage skipping through the uniform PipelineStage interface, which includes an optional canSkip method for intelligent stage evaluation before executing the run(ctx) function.

What are the built-in stages available in the default autopilot pipeline sequence?▼

The default autopilot pipeline sequence includes built-in stages for ralplan, team-exec, and ralph-verify, providing a ready-made workflow for planning, execution, and verification out of the box.

Does the pipeline orchestrator support custom stage configurations?▼

The pipeline orchestrator supports custom stage configurations by implementing the PipelineStage interface with run(ctx), allowing you to define and sequence custom stages alongside the default autopilot pipeline.