What problem does it solve? Answering broad questions about a codebase requires reading many files and tracing connections across components, which is slow and error-prone when done sequentially. This Skill decomposes a research question into parallel sub-agent investigations and synthesizes the results into a single, well-referenced research document. ## Core Features & Use Cases - Parallel sub-agent research: Spawns 2-4 concurrent Explore and general-purpose agents, each focused on a specific research dimension with file:line references. - Scope clarification: Uses structured questions to align on research depth, breadth, and focus areas before launching agents. - Structured research documents: Generates a Markdown artifact with YAML frontmatter (date, git commit, branch, topic, tags) saved under context/changes/<change-id>/research.md, with optional GitHub permalinks. - Use Case: Ask "how does the restocking classification engine work" and receive a synthesized document covering src/lib/classification.ts, related services, historical decisions from prior changes, and open questions. ## Quick Start Invoke the skill and ask a research question such as "research how velocity classification and reorder recommendations flow through the codebase".
