skills.rest
LauraFlorentin
LauraFlorentinCOMMUNITY
Explore all skills by this author →
GitHub

harness-engineer

Coordinate and maintain AI agent workflow harness environments.

1|Updated Feb 11, 2026
Search Tags:#agent#orchestration#circuit-breaker#gc#harness#onboard
One-click install
npx skills add https://github.com/LauraFlorentin/skills-marketplace --skill harness-engineer
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: harness-engineer
Source: https://github.com/LauraFlorentin/skills-marketplace/tree/main/harness-engineer
Command: npx skills add https://github.com/LauraFlorentin/skills-marketplace --skill harness-engineer

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Harness Engineer provides an end-to-end governance and scaffolding system to prevent agent loops, orient new sessions, repair failures, and manage entropy in AI agent workflows.

Core Features & Use Cases

  • Scaffold a production-ready harness (docs, features, init script, progress logs, layers, hooks)
  • Diagnose agent failures and patch the harness to prevent recurrence
  • Garbage collect drift and drifted references across features.json, docs, and constraints
  • Orient sessions automatically and map the environment for new agents

Quick Start

Install the harness-engineer plugin and run /harness-engineer:init to scaffold a complete harness for your project.

Frequently Asked Questions about harness-engineer

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

FAQPage Schema
How do I prevent infinite loops when orchestrating multi-agent workflows?▼

Multi-agent orchestration requires circuit-breaker integration and automated safety rails to prevent loops. Harness Engineer enforces these architectural constraints to sustain a reliable, auditable agent-first pipeline.

What is an agent harness and why do I need one for production?▼

An agent harness provides governance and scaffolding for AI agent workflows, preventing loops and repairing failures. You need it to maintain architectural layering and manage entropy in production agent environments.

How do I scaffold a production-ready environment for AI agent workflows?▼

Scaffolding an agent workflow environment involves generating docs, features, init scripts, progress logs, layers, and hooks. Running the init command builds a complete harness to orient new sessions and map the environment.

Can I diagnose and repair AI agent failures automatically in a harness?▼

Diagnosing agent failures and patching the harness to prevent recurrence is supported. The system applies doctor checks and drift checks to repair failures and maintain a healthy agent-first pipeline.

What's the best way to garbage collect drifted references in agent workflow files?▼

Garbage collecting drift across features.json, docs, and constraints removes outdated references. The harness gc functionality cleans drifted data to sustain an auditable, production-ready agent environment.

Does multi-agent orchestration require architectural layering to work safely?▼

Multi-agent orchestration requires architectural layering, drift checks, and automated safety rails to work safely. These constraints prevent agent loops and sustain a healthy, auditable pipeline.

Related Skills

Explore More Software Engineering Skills→
RCSnyderRCSnyder

init

Scaffold the Lights Out SWE harness into a target Git repository.

Community
Basic
ririntoririnto

jvm-gc-diagnostics

Analyzes JVM garbage-collection behavior and collector tradeoffs from GC logs and JFR data for Java LTS baselines 8-25.

Community
Advanced
chekoschekos

tunelito-start

Orient new agents by reading documentation and inspecting the project tree.

Community
Basic
sakamoto-family-smilesakamoto-family-smile

dmux-workflows

Coordinate parallel AI agent sessions across multiple harnesses with dmux.

Community
Advanced