conduct-spike

# Conduct Spike !`${CLAUDE_PLUGIN_ROOT}/bin/accelerator config context --skill conduct-spike --fail-safe` !`${CLAUDE_PLUGIN_ROOT}/bin/accelerator config agents --fail-safe` If no "Agent Names" section appears above, use these defaults: accelerator:web-search-researcher, accelerator:codebase-locator, accelerator:codebase-analyser, accelerator:codebase-pattern-finder, accelerator:documents-locator, accelerator:documents-analyser. **Work items directory**: !`${CLAUDE_PLUGIN_ROOT}/bin/accelerator config path work --fail-safe` You are tasked with conducting a **spike**: a time-boxed, uncertainty-reducing investigation. The spike is mostly conceptual, but a question is often best answered by building a small throwaway prototype or test case. Your job is to take an open, ambiguous brief and **iteratively reduce its uncertainty through collaborative discussion** — backed by agent-driven research and, where needed, small experiments — until the team can proceed with confidence, then record the outcome. This is **not** an interview and **not** a report you generate alone. It is a **collaborative discussion**: you surface evidence, propose interpretations, and the human steers — together you converge on answers. Lean on the research agents to do the legwork so the conversation stays high-level. ## Initial Response When this command is invoked: 1. **If a brief reference was provided**, resolve it in this order: - A **path** (e.g. `meta/work/0003-skill-evaluation-framework-sel...