discover

Guide problem identification, ICP drafting, and hypothesis validation for founders.

1|Updated Feb 20, 2026
One-click install
npx skills add https://github.com/october-academy/agnt --skill discover-october-academy
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: discover
Source: https://github.com/october-academy/agnt/tree/main/skills/discover
Command: npx skills add https://github.com/october-academy/agnt --skill discover-october-academy

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Guides solo founders and early-stage builders through choosing a viable customer problem, drafting an initial Ideal Customer Profile (ICP), and forming a testable hypothesis so discovery work becomes focused and actionable instead of vague and untestable.

Core Features & Use Cases

  • Structured discovery workflow: three-step interactive flow that asks targeted questions to identify and refine a problem statement.
  • ICP drafting and push-back: generates ICP candidates, validates specificity with one corrective follow-up, and captures a concise ICP draft.
  • Hypothesis & challenge: composes a monetization-style hypothesis (who would pay, for what, and how much) and optionally walks through adversarial checks to surface vulnerable assumptions.
  • Stateful integration: reads and updates local AGNT_DIR state.json, writes journey-brief.md, and queues or submits events to the agentic30 MCP for practice tracking.
  • Use case: a solo founder with no clear idea can run the flow to produce a saved problem, ICP draft, and a concrete hypothesis ready for interview validation.

Quick Start

Start the discovery flow by invoking the discover skill and answer the prompts to produce a saved problem statement, ICP draft, and hypothesis.

Frequently Asked Questions about discover

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

FAQPage Schema
How do I define an Ideal Customer Profile and a testable hypothesis for customer development?

The discovery workflow generates ICP candidates, validates their specificity with a corrective follow-up question, and captures a concise ICP draft tailored to your product problem.

What is the best way to find sellable problems during early-stage product discovery?

An adversarial challenge mechanism composes a monetization hypothesis—who pays, for what, and how much—and walks through adversarial checks to surface vulnerable assumptions in your problem selection.

Can I sync my customer development state and journey brief with an MCP server?

The workflow writes a journey-brief.md file and creates sync events for later submission, ensuring your discovery progress is saved locally and can be tracked remotely.

Does the discovery workflow support solo founders with no clear product idea?

No external dependencies are required to start; the flow operates independently by managing local state files to track your discovery progress from initial problem selection to hypothesis generation.