skills.rest
kwcantrell
kwcantrellCOMMUNITY
Explore all skills by this author →
GitHub

harness-evolve

Optimize system prompts, tools, sub-agents, and server topology for programming tasks.

Updated Jul 5, 2026
Search Tags:#optimization#runtime#AI#programming#harness
One-click install
npx skills add https://github.com/kwcantrell/rusty-agent --skill harness-evolve-kwcantrell
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: harness-evolve
Source: https://github.com/kwcantrell/rusty-agent/tree/main/.agents/skills/harness-evolve
Command: npx skills add https://github.com/kwcantrell/rusty-agent --skill harness-evolve-kwcantrell

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires agent-runtime-config, llama-server, node, npm, docker, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill provides a platform to optimize the harness (system prompt, tools, sub-agents, agent-side SDLC skills, memory, sampling, server topology) for complex programming tasks. It iteratively edits genome/config/code, evaluates against a live model, and keeps changes that improve task success and reduce token usage.

Core Features & Use Cases

  • System Prompt Tuning: Adjust system prompts to improve the performance of the runtime harness.
  • Code Optimization: Iteratively refine code and configurations to enhance the harness's efficiency.
  • Evaluation and Iteration: Evaluate harness changes against real-world tasks, and iterate based on performance results.
  • Use Case: A developer wants to optimize the harness for building a working TypeScript website. The Skill will assist in iterating through different configurations and prompts to find the most efficient and effective setup.

Quick Start

Run the harness-evolve skill to begin the optimization process for a given programming task.

Frequently Asked Questions about harness-evolve

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

FAQPage Schema
How do I optimize a runtime harness for complex programming tasks?▼

Runtime harness optimization iteratively evaluates edits to system prompts, tools, sub-agents, and sampling against a live model, retaining configurations that increase task success and reduce token usage for complex programming tasks.

What components of an AI agent runtime can be tuned for better performance?▼

Tunable runtime components include system prompts, tools, sub-agents, agent-side SDLC skills, memory, sampling parameters, and server topology. Iterating on these elements allows the harness to achieve higher task success rates and lower token consumption.

Do I need a live server to run runtime harness optimization?▼

Yes, evaluating harness optimizations requires live server access and appropriate runtime configurations. The system evaluates configuration and prompt edits by testing them dynamically against an active model to measure actual task success and token usage.

How does iterative evaluation improve an AI programming agent's efficiency?▼

Iterative evaluation improves AI programming efficiency by testing harness modifications against real-world tasks, keeping only the changes that demonstrably increase task success and reduce token usage, thereby discarding ineffective configurations automatically.

Can I use Docker and Node to configure an AI agent runtime harness?▼

Configuring an AI agent runtime harness requires Node, npm, Docker, llama-server, and agent-runtime-config dependencies. These environment tools provide the necessary infrastructure to execute live model evaluations and manage the iterative optimization process.

What is the best way to reduce token usage for complex AI programming workflows?▼

The best way to reduce token usage for complex AI programming workflows is to iteratively tune the harness components, such as system prompts and sub-agents, evaluating each change against a live model to permanently retain only efficiency-improving configurations.

Related Skills

Explore More Software Engineering Skills→
ersinkocersinkoc
56

sc-docker

Identify Docker security issues in Dockerfiles, compose configurations, and runtime containers.

Community
Advanced
vuralserhat86vuralserhat86
43

prompt_optimizer

Analyze and rewrite prompts to reduce token usage and improve success rates.

Community
Intermediate
weebcoder101weebcoder101
3

customize-opencode

Modify opencode configurations for agents, models, keybindings, and themes.

Community
Intermediate
mcauduro0mcauduro0
1

senior-prompt-engineer

Analyze and optimize LLM prompts with Python scripts.

Community
Advanced