What problem does it solve? Turning open-ended research questions into trustworthy, decision-ready artifacts is slow and error-prone: claims go uncited, sources go stale, and finished reports get reprocessed by every downstream consumer. This Skill structures the entire research lifecycle so every claim carries a source, a confidence level, and a re-check date. ## Core Features & Use Cases - Three research modes: Draft a deep-research prompt for external tools (ChatGPT, Gemini, Perplexity), Run native web research through parallel subagent fan-out, or Process a finished report into a distilled cited summary. - Typed research packs: Shipped packs for market, domain, technical, competitive, user-voice, and academic literature research, each with prioritized dimensions, source craft, freshness bars, and two-source verification classes. - Verification and lifecycle: Claims ledger with verified/disputed/unverified status, optional red-team passes, staleness maps, and Refresh/Deepen workflows that update existing run folders instead of starting over. - Use Case: Ask it to research whether to enter a new market; it holds a plan gate, fans out researcher subagents behind a research firewall, writes digests to disk as they land, and delivers a decision-first report with a source appendix and staleness map. ## Quick Start Ask the assistant to run deep recon on a topic, for example: research the competitive landscape for my product idea and produce a cited decision-ready report.
