spike

You can conduct pre-intent feasibility exploration by running focused, throwaway experiments to validate or refute an approach before writing a full implementation, generating a durable record of the hypothesis and results.

A spike is a throwaway experiment used to validate or refute feasibility hypotheses related to algorithmic, integration, or performance characteristics before committing to a specific development approach.

You run a feasibility spike by executing minimal, disposable Python experiments to test a specific hypothesis, requiring knowledge of the success criteria, relevant data, and necessary tools to generate a recommendation.

Use throwaway experiments for pre-intent analysis when an approach is uncertain and requires validation, allowing you to de-risk the strategy by assessing algorithmic or performance characteristics before formalizing an intent.

Yes, feasibility testing generates a SPIKE.md record in the dekspec/spikes directory, capturing the hypothesis, experiment results, and final recommendation for future reference.

You need knowledge of the experiment's success criteria, relevant input data, and necessary tools to run integration feasibility experiments and accurately validate or refute the approach.