two-agent-harness

Create initializer and coding agents with progress tracking for multi-session development.

9|Updated Dec 13, 2025
One-click install
npx skills add https://github.com/ingpoc/SKILLS --skill two-agent-harness
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: two-agent-harness
Source: https://github.com/ingpoc/SKILLS/tree/main/two-agent-harness
Command: npx skills add https://github.com/ingpoc/SKILLS --skill two-agent-harness

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires node, python3, git, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This skill sets up a complete two-agent development system that separates planning from implementation, enabling multi-session, recoverable workflows. It creates initializer-agent (for project planning) and coding-agent (for incremental implementation), along with enforcement hooks and progress tracking infrastructure to ensure traceable progress.

Core Features & Use Cases

  • Create an initializer-agent and a coding-agent to divide labor between planning and implementation.
  • Provide enforcement hooks, progress tracking, and reference documentation to support long-running projects.
  • Use when you need a structured handoff between planning and coding across multiple sessions, with recovery from interruptions.

Quick Start

Run the setup-two-agent-system.sh script to install the two-agent harness and initialize progress tracking.

Frequently Asked Questions about two-agent-harness

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

FAQPage Schema
How do I split project planning and coding into separate agents for long-running development?▼

A two-agent harness separates project planning from implementation by deploying an initializer-agent for planning and a coding-agent for execution. This division provides structured handoffs and progress tracking infrastructure to support recoverable multi-session development.

What is the best way to recover coding agent progress after an interruption across multiple sessions?▼

Recovering coding agent progress across multiple sessions requires a progress harness with enforcement hooks and tracking infrastructure. This setup ensures strict progress tracking, allowing incremental features to resume traceably after interruptions without losing prior context.

How do I set up a two-agent system for multi-session project development?▼

To set up a two-agent system for multi-session development, run the setup-two-agent-system.sh script. This script installs the two-agent harness, initializes progress tracking, and configures the initializer and coding agents along with enforcement hooks.

Do I need git and python3 to run a two-agent coding workflow?▼

Yes, you need git, node, and python3 installed to run a two-agent coding workflow. These dependencies provide the environment necessary for executing setup scripts, managing version control, and running the progress tracking infrastructure.

When should I use a two-agent harness instead of a single coding agent?▼

You should use a two-agent harness instead of a single coding agent when a project requires hundreds of incremental features, strict progress tracking, and recovery from interruptions across multiple sessions. It separates planning from implementation to manage complex development workflows.