harness-discovery

Clarify ambiguous project goals and constraints during early-stage planning.

12|Updated Mar 16, 2026
One-click install
npx skills add https://github.com/catu-ai/easyharness --skill harness-discovery
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: harness-discovery
Source: https://github.com/catu-ai/easyharness/tree/main/.agents/skills/harness-discovery
Command: npx skills add https://github.com/catu-ai/easyharness --skill harness-discovery

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Teams often start large or ambiguous work without a shared understanding of goals, constraints, and trade‑offs, leading to wasted effort and indecision.

Core Features & Use Cases

  • Socratic questioning to expose purpose, constraints, success criteria, and workflow direction.
  • Explorer subagents on demand for bounded repository investigations when local context is insufficient.
  • Structured discovery output that captures problem statement, key constraints, chosen direction, acceptance criteria, and the next workflow step.
  • Use Cases: Initiating a new feature in a codebase, reopening archived work, or deciding between multiple architectural approaches.

Quick Start

Ask the harness‑discovery skill to clarify your project's objectives and define the next actionable step.

Frequently Asked Questions about harness-discovery

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

FAQPage Schema
How do I turn vague project ideas into actionable plans?

To turn vague project ideas into actionable plans, apply Socratic questioning to clarify ambiguous project goals and constraints. This structured discovery process produces a concise summary and defines the next actionable workflow step.

What is the best way to clarify ambiguous project goals during early-stage planning?

The best way to clarify ambiguous project goals during early-stage planning is using Socratic questioning to expose purpose and success criteria. This method generates a structured discovery summary capturing constraints and the chosen architectural direction.

Can I use explorer subagents for repository investigations when local context is insufficient?

Yes, you can leverage optional explorer subagents for bounded repository investigations when local context is insufficient. They operate on demand to gather missing codebase context required to decide between multiple architectural approaches.

How do you decide between multiple architectural approaches in a codebase?

Deciding between multiple architectural approaches requires Socratic questioning to expose project constraints and trade-offs. This produces a structured discovery output that captures the chosen direction and defines the next workflow step.

When should I use a discovery workflow for software planning?

Use a discovery workflow for software planning when the next move is unclear or you are initiating a new feature in a codebase. It helps establish shared understanding of problem statements and acceptance criteria before large work begins.