What problem does it solve? Answering complex questions with a single web search often yields shallow, unsourced, or unreliable results. This Skill orchestrates parallel research agents that classify sources by credibility tier, cross-reference findings, and enforce hard quality gates so every claim in the final report is grounded in verifiable evidence. ## Core Features & Use Cases - Parallel Multi-Agent Research: Launches breadth, depth, and fact-checking agents that investigate distinct angles of a question simultaneously. - Source Tier Classification: Labels every source from Tier 1 (primary/official) to Tier 4 (unverifiable) and assigns confidence labels (Established, Likely, Emerging, Contested, Speculative) to each finding. - Hard Quality Gates: Enforces citation requirements, blocks fabricated URLs, discloses failed research angles, and flags reports relying too heavily on low-tier sources. - Use Case: Ask "Should we migrate from Redis to Valkey?" with thorough scope — the Skill dispatches agents to compare licensing, performance, and ecosystem, verifies single-source claims, and writes a dated report to ~/Documents/deep-research/ with an executive summary and research gaps. ## Quick Start Ask the assistant to research a topic thoroughly, for example: "Do a thorough deep research on the tradeoffs between SQLite and PostgreSQL for new web apps."
