skills.rest
LSDIPPOLLC
LSDIPPOLLCOFFICIAL
Explore all skills by this author →
GitHub

harnesscp-ergonomics

Calibrate VS Code Copilot interaction style for verbosity, autonomy, and confirmation behavior.

Updated Mar 31, 2026
Search Tags:#autonomy#trust calibration#copilot#verbosity#ergonomics#feedback memory#error communication
One-click install
npx skills add https://github.com/LSDIPPOLLC/harness-cp --skill harnesscp-ergonomics
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: harnesscp-ergonomics
Source: https://github.com/LSDIPPOLLC/harness-cp/tree/main/.github/skills/harnesscp-ergonomics
Command: npx skills add https://github.com/LSDIPPOLLC/harness-cp --skill harnesscp-ergonomics

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill fixes poor interaction habits in VS Code Copilot by aligning verbosity, autonomy, confirmation behavior, tone, and error reporting with the user’s preferred working style.

Core Features & Use Cases

  • Trust Calibration: Choose how independently the model should act, from cautious confirmation-heavy behavior to fully autonomous execution.
  • Communication Style Control: Set concise, collaborative, or teaching-oriented output so the assistant feels helpful instead of noisy or evasive.
  • Feedback Capture: Turn recurring user corrections into durable memory so the model improves across sessions.
  • Use Case: A user says the assistant is too verbose and asks it to stop over-explaining; the Skill updates interaction guidance to keep replies short, direct, and blocker-focused.

Quick Start

Ask the harnesscp-ergonomics skill to adjust Copilot to a quiet expert style with concise answers, fewer questions, and clearer error messages.

Frequently Asked Questions about harnesscp-ergonomics

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

FAQPage Schema
How do I make VS Code Copilot less verbose and stop over-explaining code?▼

To stop Copilot over-explaining, calibrate its interaction style to enforce concise output. This adjusts verbosity settings to provide short, direct answers instead of noisy over-explanations during coding sessions.

What is trust calibration for AI coding agents and how does it work?▼

Trust calibration for AI coding agents sets how independently the model acts, ranging from cautious confirmation-heavy behavior to fully autonomous execution. It adjusts autonomy levels to match your preferred working style in VS Code Copilot workflows.

Can I permanently save corrections for Copilot so it remembers my preferences across sessions?▼

Yes, you can save corrections permanently using feedback memory. This turns recurring user corrections into durable memory, allowing the model to improve and remember your preferences across multiple VS Code Copilot coding sessions.

How do I adjust the tone and communication style of my AI pair programmer?▼

You can adjust the tone of your AI pair programmer by setting communication style preferences to concise, collaborative, or teaching-oriented output. This ensures the assistant feels helpful instead of evasive or noisy during coding tasks.

Does this Copilot ergonomics calibration work with agent personalities and error reporting?▼

Yes, Copilot ergonomics calibration works with per-agent tone guidance and error communication. It updates status updates and error reporting to align with your preferred working style without performing unsafe actions in VS Code.

Related Skills

Explore More Software Engineering Skills→
fabioc-alohafabioc-aloha
1

VS Code Environment Skill

Configure VS Code extensions, workspace settings, and debug/build tasks.

Community
Intermediate
fmflurryfmflurry
171

learned-user-correction-integration-pattern

Extract recurring user corrections into reusable rules and apply them across sessions.

Community
Intermediate
AVIDS2AVIDS2
618

memcode

Maintain a shared persistent memory layer for AI coding agents.

Community
Advanced
smriksmrik
1

Output Styles Guide

Modifies Claude Code system prompts for non-engineering workflows.

Community
Intermediate