workflow:harness

Automate iterative development cycles by orchestrating Claude Code in a bash loop.

3|Updated Dec 30, 2025

Search Tags:#acceptance criteria #iterations #autonomous #specs #harness #loop #PLAN.md

Ready to run

npx skills add https://github.com/guicheffer/devorch-cli --skill workflow-harness

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: workflow:harness
Source: https://github.com/guicheffer/devorch-cli/tree/main/templates/skills/workflow/harness
Command: npx skills add https://github.com/guicheffer/devorch-cli --skill workflow-harness

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates complex development tasks by running Claude Code in an external loop, allowing for iterative progress tracking and autonomous execution based on defined specifications.

Core Features & Use Cases

Autonomous Development: Enables AI agents to work through multi-step features without manual intervention.
Progress Tracking: Utilizes acceptance criteria in PLAN.md to monitor task completion.
Iterative Refinement: Generates detailed logs for each iteration, providing a clear history of changes and approaches.
Use Case: Implement a new authentication system by initializing a feature, generating detailed specs, defining tasks with acceptance criteria, and then letting the harness loop autonomously build and refine the feature until all criteria are met.

Quick Start

Initialize a new feature folder named 'user-profile' using the devorch harness init command.

Frequently Asked Questions about workflow:harness

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

FAQPage Schema

How do I automate iterative development cycles for complex features using Claude Code?▼

Automate iterative development cycles by running Claude Code in an external bash loop that orchestrates task breakdown, tracks progress via acceptance criteria in PLAN.md, and generates detailed iteration logs for autonomous execution.

What is autonomous spec-based progress tracking and how does it work?▼

Spec-based progress tracking monitors task completion by evaluating defined acceptance criteria in a PLAN.md file, allowing an external bash loop to autonomously determine if iterative feature implementation and refactoring steps meet the required specifications.

Can I use an external bash loop to manage multi-step feature implementation without manual intervention?▼

Yes, you can use an external bash loop to manage multi-step feature implementation, enabling AI agents to autonomously work through complex tasks and refactoring by orchestrating execution and tracking progress against defined acceptance criteria.

How do I start an autonomous development workflow for a new feature like user authentication?▼

Initialize a new feature folder using the harness init command, generate detailed specs, define tasks with acceptance criteria in PLAN.md, and let the external loop autonomously build and refine the feature until all criteria are met.

Does autonomous AI development require predefined acceptance criteria to function properly?▼

Yes, predefined acceptance criteria are required to monitor task completion and guide the external bash loop, ensuring the autonomous development workflow builds and refines features accurately until all specified conditions are met.

What are the limitations of running autonomous AI agents for complex refactoring tasks?▼

Autonomous AI agents rely entirely on the accuracy of the initial specs and acceptance criteria in PLAN.md; poorly defined criteria or ambiguous task breakdowns can cause the external bash loop to iterate inefficiently or fail to meet feature requirements.