pipeline

Chain agents into sequential or branching pipelines with stateful context passing.

28|5|Updated Jan 26, 2026
One-click install
npx skills add https://github.com/MeroZemory/oh-my-droid --skill pipeline-merozemory
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: pipeline
Source: https://github.com/MeroZemory/oh-my-droid/tree/main/skills/pipeline
Command: npx skills add https://github.com/MeroZemory/oh-my-droid --skill pipeline-merozemory

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill turns disparate agent tasks into cohesive workflows so that complex requests can flow from exploration through planning to execution without manual handoffs or repeated context building.

Core Features & Use Cases

  • Sequential and branching pipelines let you route outputs through architect, executor, designer, or critic agents based on the task complexity or content.
  • Stateful context passing and retry logic guarantees each stage receives structured information, supports fallbacks, and recovers from transient failures.
  • Use Case: Run the review preset to scan architecture, analyze implications, critique the result, and implement changes across multiple specialists in one command.

Quick Start

Say '/pipeline review add rate limiting to API' to run the review pipeline for that feature request.

Frequently Asked Questions about pipeline

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

FAQPage Schema
How do I chain multiple agents into a sequential workflow?

You can chain agents into sequential or branching pipelines to distribute complex tasks across specialized runtimes. This coordinates architecture reviews, implementation plans, and debugging sessions across multiple agents with structured context passing.

How does context passing work in agent pipelines?

Context passing in agent pipelines ensures each stage receives structured information by managing state across sequential tasks. It guarantees downstream agents inherit the exact context needed, supporting fallbacks and recovery from transient failures.

Can I create branching decisions for different agent tasks?

Yes, you can create branching decisions to route outputs through architect, executor, designer, or critic agents based on task complexity or content. This allows dynamic workflow orchestration tailored to the specific requirements of each stage.

What is the best way to automate architecture reviews across multiple agents?

The best way to automate architecture reviews is using a preset pipeline command that scans architecture, analyzes implications, critiques results, and implements changes across multiple specialists in one command.

Does workflow orchestration handle retries and state management automatically?

Workflow orchestration handles retries and state management automatically to ensure reliable completion of complex tasks. It supports fallbacks and recovers from transient failures during sequential or branching agent pipelines.

When should I use branching pipelines instead of sequential tasks?

Use branching pipelines instead of sequential tasks when you need to route outputs to different specialized agents based on content analysis or task complexity. Sequential pipelines are better suited for straightforward, step-by-step execution flows.