spike

Decompose ideas into feasibility questions and execute quick spike experiments.

78|16|Updated Apr 23, 2026
One-click install
npx skills add https://github.com/sheawinkler/hermes-agent-ultra --skill spike-sheawinkler
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: spike
Source: https://github.com/sheawinkler/hermes-agent-ultra/tree/main/skills/software-development/spike
Command: npx skills add https://github.com/sheawinkler/hermes-agent-ultra --skill spike-sheawinkler

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Spike solves the problem of uncertainty in early ideas by enabling quick, disposable experiments to test feasibility before committing to a full build.

Core Features & Use Cases

  • Structured 5-step spike workflow (decompose → align → research → build → verdict) to isolate feasibility questions
  • Lightweight, standalone experiments that can be thrown away after learning
  • Clear verdicts (VALIDATED, PARTIAL, INVALIDATED) to guide real development decisions

Quick Start

Decompose a user idea into 2–5 spike questions, run a minimal experiment, and record a verdict.

Frequently Asked Questions about spike

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

FAQPage Schema
What is a spike experiment for validating software feasibility?

To run a spike, decompose a user idea into 2 to 5 feasibility questions, execute a minimal throwaway experiment to research unknowns, and record a formal verdict of VALIDATED, PARTIAL, or INVALIDATED to guide development decisions.

When should I run throwaway experiments instead of building a prototype?

Run a spike to test idea feasibility by breaking it into discrete questions, executing minimal experiments, and capturing learnings with a formal verdict to guide real development decisions without committing to a full build.

How do I evaluate if a new feature idea is technically feasible?

Run a spike to test idea feasibility by breaking it into discrete questions, executing minimal experiments, and capturing learnings with a formal verdict to guide real development decisions without committing to a full build.

Can I use lightweight experiments to test unknowns in early product planning?

Run a spike to test idea feasibility by breaking it into discrete questions, executing minimal experiments, and capturing learnings with a formal verdict to guide real development decisions without committing to a full build.

What is the best way to document findings from a technical spike?

Run a spike to test idea feasibility by breaking it into discrete questions, executing minimal experiments, and capturing learnings with a formal verdict to guide real development decisions without committing to a full build.