What problem does it solve? Answering complex questions from memory alone risks hallucinated facts and missing sources. This Skill runs a structured, evidence-grounded research protocol that decomposes a question, retrieves real web sources, verifies every claim, and delivers a cited report. ## Core Features & Use Cases - Iterative Retrieval Loop: Decomposes the question into subtopics and runs up to 4 rounds of web_search and web_fetch, maintaining an evidence ledger keyed by subtopic. - Parallel Synthesis & Verification: Dispatches parallel sme workers to synthesize findings, then two reviewers verify each claim against its cited evidence, and a critic challenges high-stakes claims. - Cited Report Output: Produces a report or brief with per-subtopic sections, source citations, honest disagreement surfacing, and explicit UNVERIFIED markers for ungrounded topics. - Use Case: Ask "What are the current best practices for RAG chunking strategies?" and receive a fact-checked report where every load-bearing claim links to a fetched primary source. ## Quick Start Ask the agent to run deep research on your question, for example: run deep research on the trade-offs between vector databases for self-hosted RAG systems.
