What problem does it solve? It answers feasibility questions that research alone cannot resolve by building small disposable prototypes, so you avoid committing to an approach that turns out not to work. ## Core Features & Use Cases - Decompose and Prioritize: Break an idea into 2-5 independent feasibility questions framed as Given/When/Then, ordered by risk so the idea-killing spike runs first. - Build and Compare: Create one standalone directory per spike with runnable CLI, HTML page, or test output, including parallel comparison spikes (e.g., pdfjs vs camelot) with head-to-head verdicts. - Verdict Reporting: Close each spike with a VALIDATED, PARTIAL, or INVALIDATED verdict documenting what worked, surprises, and recommendations for the real build. - Use Case: Before building a streaming chat feature, spike websocket token streaming to confirm the client receives chunks under 100ms, then throw the code away. ## Quick Start Ask the agent to spike whether your chosen library can stream LLM tokens over a websocket before you commit to the full implementation.
