spike

Decompose ambiguous ideas into risk-ordered Given/When/Then feasibility spikes with verdicts.

19|4|Updated Apr 22, 2026
One-click install
npx skills add https://github.com/carterwayneskhizeine/hermes-agent-windows-R --skill spike-carterwayneskhizeine
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: spike
Source: https://github.com/carterwayneskhizeine/hermes-agent-windows-R/tree/main/skills/software-development/spike
Command: npx skills add https://github.com/carterwayneskhizeine/hermes-agent-windows-R --skill spike-carterwayneskhizeine

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you de-risk product or engineering decisions by turning an idea into a quick, throwaway experiment that produces observable evidence before you commit to real development.

Core Features & Use Cases

  • Feasibility-first spike planning: Breaks the idea into 2–5 independent feasibility questions and orders them by risk using Given/When/Then framing.
  • Research-light but decisions-ready: Performs just enough research per spike to choose an approach, compare credible alternatives, and avoid building blind.
  • Interactive, verifiable outputs: Biases toward runnable CLIs, minimal HTML demos, or small servers (or tests) so the results are felt and evidenced.
  • Clear verdicts for next steps: Produces a structured VALIDATED | PARTIAL | INVALIDATED outcome and recommends what to do in the real build.

Quick Start

Ask: "Spike this idea for me: [your goal]. Break it into 2–5 feasibility questions, research just enough to choose an approach, build minimal runnable spikes, and end with a VALIDATED/PARTIAL/INVALIDATED verdict and recommendation."

Frequently Asked Questions about spike

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

FAQPage Schema
How do I run a feasibility spike to validate an idea before building?

To run a feasibility spike, decompose your idea into 2-5 risk-ordered Given/When/Then questions, perform targeted research, build minimal runnable artifacts, and produce a VALIDATED, PARTIAL, or INVALIDATED verdict with build recommendations.

What is a spike in software engineering decision-making?

A spike is a short, disposable feasibility experiment that de-risks software and product decisions by turning ambiguous ideas into observable evidence, comparing approaches, and proving unknowns before committing to production investment.

What's the best way to de-risk early-stage product planning and approach comparison?

The best way to de-risk early-stage planning is executing research-light spikes that prioritize credible alternatives, build interactive verifiable outputs like minimal HTML demos or small servers, and end with structured verdicts.

Can I build interactive prototypes for build verification without full development?

Yes, you can build interactive prototypes for build verification by biasing outputs toward runnable CLIs, minimal HTML demos, or small tests, ensuring results are felt and evidenced without full production investment.

When do I need a feasibility experiment for MVP prototype validation?

You need a feasibility experiment when facing ambiguous ideas or unknowns in MVP prototype validation, requiring targeted research and minimal interactive artifacts to choose an approach and avoid building blind.

What are the limitations of using throwaway spikes for product decisions?

Throwaway spikes are limited to early-stage planning and approach comparison, producing research-light evidence rather than production-ready code, meaning they should be discarded after generating a VALIDATED, PARTIAL, or INVALIDATED verdict.