harness-command

Create and manage custom slash command entities in an agent-harness workspace.

12|Updated Feb 6, 2026
One-click install
npx skills add https://github.com/madebywild/agent-harness --skill harness-command
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: harness-command
Source: https://github.com/madebywild/agent-harness/tree/main/.claude/skills/harness-command
Command: npx skills add https://github.com/madebywild/agent-harness --skill harness-command

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Slash commands provide a structured way to trigger one-shot actions directly from chat tools, enabling explicit manual workflows within an agent-harness workspace.

Core Features & Use Cases

  • Scaffold and organize command definitions stored under .harness/src/commands.
  • Generate per-provider outputs for Claude and Copilot with optional per-command overrides.
  • Ensure consistent frontmatter, body structure, and step-by-step execution guidance across projects.

Quick Start

Scaffold a new command with harness add command <command-id> and customize its description and steps.

Frequently Asked Questions about harness-command

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

FAQPage Schema
How do I create reusable slash commands for Claude or Copilot?

You can create custom slash commands by scaffolding definitions with a tool that stores them under .harness/src/commands and generates per-provider outputs for Claude and Copilot.

What's the best way to organize custom agent commands across different projects?

Organizing custom agent commands requires enforcing a standardized source format with consistent frontmatter and body structure under .harness/src/commands to maintain cross-project consistency.

Can I apply per-command overrides when generating slash commands for multiple chat providers?

Yes, you can generate per-provider outputs for Claude and Copilot while applying optional per-command overrides to customize specific slash command behaviors individually.

How do I trigger one-shot manual actions directly from chat tools?

Slash commands provide a structured way to trigger one-shot actions directly from chat tools, enabling explicit manual workflows within an agent-harness workspace environment.

Does harness-command work with both Claude and Copilot agent workflows?

Yes, harness-command supports generating standardized slash command outputs tailored for both Claude and Copilot environments within a shared agent-harness workspace.