What problem does it solve? Broad or ambiguous research questions often get shallow answers from a single-pass investigation. This Skill coordinates multiple parallel research agents across official documentation, codebase, and community sources, then merges findings into a deduplicated claim registry with conflict resolution and a completeness score. ## Core Features & Use Cases - Multi-Agent Parallel Research: Dispatches 2-3 background agents sharded by source type (official docs, code, community) with a web tool fallback cascade. - Claim Registry & Scoring: Normalizes, deduplicates, and cross-verifies findings, then computes a 4-signal completeness score with confidence caps. - Conditional Validation & Debate: Triggers a validator micro-loop or escalates unresolved conflicts to a structured debate only when evidence is weak. - Use Case: Ask "Should we use Redis or PostgreSQL for caching?" and receive a report with consensus claims, divergence analysis, coverage matrix, and an evidence-backed recommendation. ## Quick Start Ask the agent to run deep research on your topic, for example: research the best patterns for multi-agent orchestration with a medium budget.
