pipeline

Orchestrate multi-agent software development through design, critique, build, and review loops.

Updated Jun 12, 2026

Search Tags:#workflow #tdd #orchestration #pipeline #agentic #autonomous

Ready to run

npx skills add https://github.com/ambrovia/agent-skills-pipeline --skill pipeline-ambrovia

💬 Prompt for ChatGPT / Claude Chat▼

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the lack of structure in agentic coding by enforcing a rigorous, multi-persona workflow that separates planning, design, building, and review to ensure high-quality, reliable software delivery.

Core Features & Use Cases

Autonomous Orchestration: Drives work packages through a complete lifecycle (design, critique, build, review, retro, ship) without human intervention.
Persona-Based Quality Control: Enforces producer/evaluator separation by using distinct agents for planning and reviewing, preventing self-grading bias.
Use Case: Use this to autonomously implement a batch of backend work packages, ensuring each passes TDD, architectural review, and CI verification before merging.

Quick Start

Invoke the pipeline skill by providing the specific work package IDs you want to drive to completion.

Frequently Asked Questions about pipeline

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

FAQPage Schema

How do I orchestrate autonomous software development workflows with multi-agent personas?▼

Autonomous software development workflows are orchestrated by managing multi-agent personas through a structured design-critique-build-review loop, ensuring strict adherence to acceptance criteria and architectural standards without requiring human intervention.

What is producer-evaluator separation in agentic coding pipelines?▼

Producer-evaluator separation prevents self-grading bias by using distinct agents for planning and reviewing, ensuring high-quality software delivery through strict quality control during the build and review phases.

How do I enforce TDD and CI verification gates in autonomous coding pipelines?▼

TDD and CI verification gates are enforced by driving work packages through a complete lifecycle that includes automated verification and architectural review, ensuring each package passes before merging.

Can I use autonomous agentic workflows for complex engineering tasks requiring persistent state management?▼

Yes, autonomous agentic workflows apply to complex engineering tasks requiring persistent state management via worktrees, ensuring integrated CI/CD pipeline completion and strict adherence to acceptance criteria.

What is the best way to structure agentic coding to avoid self-grading bias?▼

The best way to avoid self-grading bias is enforcing a rigorous multi-persona workflow that separates planning, design, building, and review, driving work through design, critique, build, review, retro, and ship phases.

Do I need specific work package IDs to start an autonomous development pipeline?▼

Yes, you need to provide specific work package IDs to invoke the pipeline skill and autonomously drive those work packages to completion through the structured lifecycle.