harness-executor

Execute validated development tasks through an eight-stage harness loop.

1|Updated Apr 14, 2026
One-click install
npx skills add https://github.com/Lzhtommy/oh-my-harness-skills --skill harness-executor
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: harness-executor
Source: https://github.com/Lzhtommy/oh-my-harness-skills/tree/main/harness-executor
Command: npx skills add https://github.com/Lzhtommy/oh-my-harness-skills --skill harness-executor

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires python3, timeout, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill enables teams to run development tasks inside a bootstrapped harness environment with guaranteed validation, traceability, and memory of outcomes, reducing manual handoffs and rework.

Core Features & Use Cases

  • 8-stage execution loop: pre-validate, execute, validate, trace, memory, compile, yield, and checkpoint for disciplined task progress.
  • Plan management & delegation: curate execution plans, handle active/completed states, and delegate sub-tasks to sub-agents when needed.
  • Session handoff & recovery: preserve session context, resume from checkpoints, and maintain robust failure traces for auditing.

Quick Start

Bootstrapped harness projects can execute development tasks with validation and tracing.

Frequently Asked Questions about harness-executor

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

FAQPage Schema
How do I enforce validation and traceability in multi-step development tasks?

Multi-step development tasks require an 8-stage execution loop—pre-validate, execute, validate, trace, memory, compile, yield, checkpoint—to enforce strict guardrails, preserve session context, and reduce manual rework across complex workstreams.

What is the best way to coordinate sub-agent delegation in a harness-backed project?

Harness-backed projects coordinate sub-agent delegation through active and completed plan management, curating execution plans and delegating sub-tasks to sub-agents while maintaining robust failure traces for auditing.

How does session handoff and recovery work for interrupted automation tasks?

Session handoff and recovery preserve session context and resume interrupted automation tasks from checkpoints, maintaining robust failure traces to audit outcomes and ensure reliable execution continuity.

Do I need a bootstrapped harness environment to run validated execution loops?

Yes, running validated execution loops requires a bootstrapped harness environment to apply strict validation, memory, and traceability across complex development workstreams with robust failure handling.

Can I manage execution plans and checkpoint states for complex workstreams?

Yes, plan management handles active and completed states for complex workstreams, allowing you to curate execution plans, checkpoint progress, and yield validated outcomes with end-to-end traceability.

Why does task execution fail without pre-validation and memory stages?

Task execution fails without pre-validation and memory stages because the harness execution loop relies on these stages to enforce guardrails, trace outcomes, and maintain session context across multi-step development workstreams.