agent-teams

Orchestrate an inter-agent development team through a strict TDD pipeline.

Updated Apr 21, 2026

Search Tags:#automation #tdd #multi-agent #code-review #orchestrate #team-management

Ready to run

npx skills add https://github.com/joshuakessell/cornerstone --skill agent-teams-joshuakessell

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: agent-teams
Source: https://github.com/joshuakessell/cornerstone/tree/main/.claude/skills/agent-teams
Command: npx skills add https://github.com/joshuakessell/cornerstone --skill agent-teams-joshuakessell

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It provides an automated, team-based development workflow that coordinates multiple AI agents to enforce a strict TDD pipeline, reducing coordination overhead and human error.

Core Features & Use Cases

Default team roster with permanent and dynamic agents
Immutable 10-step feature task chain with dependencies
Cross-feature serialization and PR management via Merger Agent
iCPG integration for context-aware task planning and risk assessment

Quick Start

Initialize a CLAUDE project and let the Team Lead spawn feature agents to begin the TDD pipeline.

Frequently Asked Questions about agent-teams

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

FAQPage Schema

How do I orchestrate multiple AI agents for test-driven development?▼

Orchestrating multiple AI agents for test-driven development involves a Team Lead that spawns feature agents to execute a strict 10-step TDD pipeline. This enforces dependency-driven task chains and manages pull requests through a dedicated Merger Agent.

What is a multi-agent TDD pipeline and how does it work?▼

A multi-agent TDD pipeline is an automated workflow where a team roster of permanent and dynamic agents implements features through an immutable 10-step task chain. It applies dependency-driven execution and uses a dedicated Merger Agent for cross-feature serialization and PR management.

How do I enforce strict code review and role separation across automated feature branches?▼

Enforcing strict code review and role separation across automated feature branches requires a dedicated Merger Agent to manage PRs while feature agents handle implementation. This provides auditability and scalable collaboration across features without human coordination overhead.

Do I need a specific project setup to start a multi-agent TDD workflow?▼

You need to initialize a CLAUDE project to start a multi-agent TDD workflow. Once initialized, the Team Lead automatically spawns feature agents to begin executing the strict TDD pipeline and managing the dependency-driven task chains.

Can I automate context-aware task planning and risk assessment for my development team?▼

Automating context-aware task planning and risk assessment is possible through iCPG integration. This is built into the multi-agent orchestration workflow to help the Team Lead evaluate features before spawning agents and executing the TDD task chain.

What are the limitations of automating team management with a multi-agent TDD pipeline?▼

Automating team management with a multi-agent TDD pipeline relies on an immutable 10-step task chain, making the process rigid and unsuitable for ad-hoc development. It is built for projects needing strict auditability and role separation rather than flexible or exploratory coding workflows.