spike

Decompose uncertain software ideas into feasibility questions and build disposable prototypes.

Updated May 15, 2026
One-click install
npx skills add https://github.com/cabezno/bmb-encover-agent --skill spike-cabezno
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: spike
Source: https://github.com/cabezno/bmb-encover-agent/tree/main/skills/software-development/spike
Command: npx skills add https://github.com/cabezno/bmb-encover-agent --skill spike-cabezno

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you de-risk software decisions by turning an uncertain idea into concrete evidence, so you can avoid wasting time building the wrong thing.

Core Features & Use Cases

  • Feasibility decomposition: Breaks a goal into 2–5 independent, observable questions so the highest-risk unknowns are tested first.
  • Targeted research and approach selection: Surfaces competing options, reads real docs, and chooses what to prototype based on evidence.
  • Disposable builds with verdicts: Produces throwaway prototypes that end with a clear VALIDATED | PARTIAL | INVALIDATED recommendation for the real build.
  • Comparison spikes: Runs parallel variants (e.g., approach A vs B), then delivers a head-to-head winner for the decision.

Quick Start

Use the spike skill to test whether your proposed solution is feasible by asking it to decompose the idea into high-risk feasibility questions, research competing approaches, build a minimal interactive prototype, and conclude with a VALIDATED, PARTIAL, or INVALIDATED verdict.

Frequently Asked Questions about spike

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

FAQPage Schema
How do I validate the feasibility of a software idea before committing to a full build?

A technical spike validates feasibility by decomposing an uncertain idea into high-risk questions, researching approaches, building a throwaway prototype, and delivering a clear VALIDATED or INVALIDATED verdict. This evidence-driven process prevents wasted effort building the wrong solution.

What is the best way to compare two technical approaches for a web integration project?

The best way to compare technical approaches is running a comparison spike that builds parallel variants, tests observable output, and delivers a head-to-head winner. This method uses evidence-driven recommendations to decide which integration approach works best.

How do I identify and test the highest-risk unknowns in a proposed technical architecture?

Identify high-risk unknowns through feasibility decomposition, which breaks a goal into 2–5 independent, observable questions ordered by risk. Testing these unknowns first ensures the most critical technical research is validated before building the full system.

When should I use a disposable prototype instead of building production-ready code?

Use a disposable prototype when facing uncertain technical feasibility, such as web parsing or integration-heavy tasks where observable output matters. It produces throwaway builds that end with a clear PARTIAL or INVALIDATED recommendation, guiding whether the real build should proceed.

Can I use this approach to test parsing and integration feasibility for unknown data formats?

Yes, this approach applies to parsing and integration-heavy tasks where observable output matters. It researches real docs, targets feasibility questions, and builds interactive prototypes to validate whether the proposed parsing or integration solution works before implementation.