What problem does it solve?
Deep research usually ends in scattered notes, lost sources, or a summary that only makes sense to the person who did the reading. This Skill turns a research question into a permanent knowledge asset: every primary source archived verbatim, one summary per source, and a single self-contained compendium page that a cold reader can follow from sentence one.
Core Features & Use Cases
- Four-phase pipeline: find everything (brain-first, then web), archive each primary source verbatim, write a strict 1:1 summary per source, then synthesize one compendium page under
research/<topic-slug>/.
- Depth dial: five idempotent levels from base synthesis up to exhaustive saturation, where each level only adds missing work based on the index ledger.
- Quality gates: cold-read validation, a mechanical fact-check claims ledger at higher levels, and a cross-modal eval requiring every dimension to score at least 7.
- Safety and privacy: fetched content is treated as untrusted data with prompt-injection flagging, and verbatim archiving is gated by the user's retention posture.
- Use Case: Ask for a definitive guide to spaced repetition and receive an archived corpus of 15-30 sources, per-source summaries with real numbers, and a compendium with TL;DR, evidence, playbook, and pitfalls sections.
Quick Start
Ask the agent to research everything about a topic, archive the sources, summarize each one, and write a compendium.
