Validate technical assumptions through time-boxed spike experiments.
npx skills add https://github.com/uta2000/feature-flow --skill spike-uta2000
Please help me install this Agent Skill. Skill: spike Source: https://github.com/uta2000/feature-flow/tree/main/skills/spike Command: npx skills add https://github.com/uta2000/feature-flow --skill spike-uta2000
Run time-boxed technical experiments to de-risk unknowns before committing to a design or implementation plan. A spike answers the question: "Will this actually work?"
Identify the riskiest assumptions, then run a time-boxed spike to validate them before committing to design.
High-intent search queries and answers about installing and using this skill.
You can validate technical assumptions by running time-boxed spikes that design minimal experiments to test risky unknowns. This process identifies assumptions, executes targeted experiments, and delivers evidence-based verdicts to inform your design decisions before implementation.
A time-boxed spike is a focused technical experiment used to de-risk uncertain outcomes before committing to implementation. It answers whether a specific approach will work by designing minimal experiments, executing them in parallel worktrees, and compiling evidence into a final spike report.
To evaluate external API and library feasibility, run structured technical spikes that identify risky assumptions and design minimal experiments. This validates integration outcomes through evidence-based verdicts, ensuring uncertain external dependencies are tested before full design commitment.
The best way to run proof-of-concept experiments is using a structured spike workflow that identifies assumptions, designs minimal tests, and executes them in parallel worktrees. This delivers evidence-based verdicts and a concise spike report to inform your design decisions.
You should run a spike during design brainstorming or concept reviews when evaluating uncertain outcomes like external APIs, libraries, or integrations. Running time-boxed experiments de-risks these unknowns by delivering evidence-based verdicts before you commit to a design.
Spikes are limited by their time-boxed nature, meaning they provide minimal experiments rather than full implementations. They focus strictly on validating or denying technical assumptions to deliver evidence-based verdicts, so they should not be treated as production-ready code or comprehensive design documents.
imvanzenInspect repository rules and analyze codebase patterns before code changes.
JamesFincherDecompose proposals into feasibility questions and run disposable implementation spikes.
RheinmirCreate Git worktrees and tmux sessions for isolated development environments.
vibbsBuild time-boxed MVP prototypes from company.config.yaml and user-defined objectives.
