skills.rest
dformoso
dformosoCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

pick-harness-shape

Design LLM/agent system harness structure, substrate, topology, and tool configuration.

Updated Apr 28, 2026
Search Tags:#security#verification#memory management#observability#topology#LLM/agent system#harness design#substrate selection#tool layer#model choice
One-click install
npx skills add https://github.com/dformoso/positronic --skill pick-harness-shape
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: pick-harness-shape
Source: https://github.com/dformoso/positronic/tree/main/skills/defining/pick-harness-shape
Command: npx skills add https://github.com/dformoso/positronic --skill pick-harness-shape

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill helps design the structure and capabilities of a harness for LLM/agent systems, ensuring reliability and efficiency beyond conversational use.

Core Features & Use Cases

  • Custom Harness Decision: Determines if a custom harness is needed based on the project's requirements.
  • Substrate Selection: Chooses the appropriate substrate for the harness, such as Claude Agent SDK, OpenAI Agents SDK, or Google ADK.
  • Topology Design: Defines the topology, including the division of labor and improvement/recovery loop.
  • Memory & State Management: Manages how memory and state are handled within the system.
  • Tool Layer Configuration: Configures the tool layer, including the number of tools, MCP server, and permissions.
  • Verification & Recovery: Sets up verification gates, recovery strategies, and budgeting for the system.
  • Security & Trust Boundaries: Establishes security measures and trust boundaries for the system.
  • Model Choice & Sampling: Selects the appropriate model tier, temperature, reasoning effort, and token limits for each stage.
  • Observability & Evaluation: Defines the posture for trajectory instrumentation, success signals, harness-level eval, and production monitoring.
  • Artifact Creation: Generates a comprehensive artifact detailing all design decisions.
  • Hand-off & Next Steps: Guides the user through the hand-off process and prompts for further actions.
  • Use Case: Ideal for software engineers and product managers designing complex LLM/agent systems that require high reliability and performance.

Quick Start

Run the skill to start designing the harness for your LLM/agent system.

Frequently Asked Questions about pick-harness-shape

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

FAQPage Schema
How do I design a harness for an LLM agent system?▼

To design an LLM agent system harness, you must define the topology, memory management, tool layer, and verification gates to ensure reliability beyond conversational use. This process structures the system's capabilities and recovery loops.

What is the best way to select a substrate for an LLM agent harness?▼

Selecting an LLM agent harness substrate involves choosing an appropriate foundation like Claude Agent SDK, OpenAI Agents SDK, or Google ADK based on your project requirements. The right substrate ensures system efficiency and reliability.

Do I need a custom harness for my LLM agent system?▼

You need a custom LLM agent harness when your project requires specific topology designs, memory state management, and tool layer configurations that exceed standard conversational capabilities. A custom decision process evaluates these reliability needs.

How do you configure verification gates and recovery loops in agent topology design?▼

Configuring verification gates and recovery loops in agent topology design involves setting up budgeting strategies and trust boundaries to manage system failures. This ensures the LLM agent system recovers efficiently and maintains operational security.

How to choose the right model tier and token limits for LLM agent stages?▼

Choosing the right model tier and token limits for LLM agent stages requires defining the temperature, reasoning effort, and sampling parameters for each operational phase. This configuration optimizes the system's overall performance and cost budgeting.

What observability and evaluation posture is needed for LLM agent systems?▼

An effective observability and evaluation posture for LLM agent systems requires trajectory instrumentation, success signal monitoring, and harness-level evaluation. This setup ensures continuous production monitoring and reliable system performance.

Related Skills

Explore More Software Engineering Skills→
AesopScottAesopScott

budget-check

Validate budget constraints before dispatching tasks in the Cost harness.

Community
Advanced
oornneryoornnery

building-agents

Design and implement tool-using Python agents for LLM-driven reasoning.

Community
Advanced
yf8578yf8578
1

mcp-builder

Guide developers in creating MCP servers for LLM integration.

Community
Advanced
markwang2658markwang2658
31

slime-rl-training

Plan and execute LLM RL post-training with the slime Megatron+SGLang framework.

Community
Advanced