Deep Research
Deep Research Kit
For anyone who wants to understand things properly. These prompts help you steel-man opposing views, map the source landscape, apply mental models, audit your own assumptions, synthesize conflicting information, and reason from first principles.
6 promptsFree to use
ResearchThinkingAnalysis
01Steel Man
I'm about to argue that [your position]. Before I do, I need the strongest possible version of the opposing view — not a strawman, the real thing. 1. Steel-man the opposition: make their argument so strong that it genuinely makes me reconsider 2. Give me the 3 most powerful counterarguments to my position 3. Tell me which counterargument is hardest to rebut — and why Then tell me: after all this, do you think my original position still holds? What would need to be true for it to be wrong?
02Source Map
I'm researching [topic] and want to actually understand it, not just collect citations. Build me a reading map: Tier 1 — Must read first (foundational, high-signal) Tier 2 — Go deep if interested (fills in the picture) Tier 3 — Only if I need to cite something (academic, specialized) For each source: what it covers, why it matters, and what lens to read it through. Also flag: anything that's commonly cited but actually outdated, wrong, or misrepresented — I want to know before I repeat the mistake.
03Mental Model Apply
Apply the [mental model or framework — e.g., second-order thinking / Overton Window / OODA loop / Inversion] to [situation, problem, or decision I'm facing]. Walk me through step by step: 1. What does this model say I should look at first? 2. What does it reveal that I would have missed? 3. What's the non-obvious implication? 4. Where does the model break down or not quite fit this situation? End with: the one action or shift in thinking this analysis points to.
04Assumption Audit
Here's a plan / argument / belief I hold: [paste it] Do a full assumption audit: 1. List every assumption embedded in this — including the ones I take for granted 2. For each assumption, rate it: well-supported / uncertain / probably wrong 3. Identify the load-bearing assumptions — the ones that, if wrong, collapse the whole thing 4. For the 3 riskiest assumptions: what evidence would confirm or refute each one? Tell me where I'm on shakiest ground.
05Synthesis
I've been reading about [topic]. Here's what I've gathered: [paste notes, excerpts, or summaries from your sources] Synthesize this into a structured mental model: 1. The core claim all sources agree on 2. The legitimate disagreements — where smart people actually differ and why 3. The unresolved questions — what nobody knows yet 4. The common misconception that keeps appearing End with: a 3-sentence summary of [topic] I could explain to someone who knows nothing about it — and have them actually understand it.
06First Principles
Break down [complex topic, decision, or problem] from first principles. Don't start from what everyone already believes — start from what we can actually know with certainty. 1. What are the bedrock facts or physical constraints? (Things that can't be argued away) 2. What do we reliably know from evidence? 3. What's inferred or assumed on top of that? 4. Where does the real uncertainty live — and is it uncertainty about facts or about values? End with: what changes if I think about this from first principles instead of starting from the conventional view?
More packs on the way
I drop new prompt packs on Facebook and Instagram regularly. Follow to get them as soon as they land.