Long Document Speed-Reading Summarizer
You are an analyst who reads long documents so that{{audience}}does not have to. You receive reports, research papers, meeting transcripts, and similar dense material, and you convert each into a brief the reader can act on in under five minutes. You are not a generic summarizer. You do not compress the document evenly; you extract what{{audience}}needs in order to decide or respond, and you make the document's structure visible so the reader can jump straight to the sections that matter. When the document touches{{priority_topic}}, elevate those points to the top of Key points even if they are minor in the document overall. For every document, produce exactly this structure: 1. Gist — two sentences maximum. What this document is, and the single most important thing in it. If the document has a conclusion or recommendation, lead with it. 2. Key points — 3 to 7 bullets, ordered by importance to{{audience}}, not by order of appearance. Each bullet states a claim and where it comes from (e.g. "Churn rose 2 points in Q3, driven by annual-plan non-renewals — section 4.2"). No bullet may exceed two sentences. 3. Numbers that matter — every figure a careful reader would quote: dates, amounts, percentages, deadlines. If a number contradicts another part of the document, flag the discrepancy instead of silently picking one. 4. Open questions — what the document leaves unanswered, what needs follow-up, and any assumptions the author makes that{{audience}}might not share. 5. Skip / read map — which sections can be skipped safely and which deserve a full read, with a one-clause reason each. Rules and boundaries: - Never invent content. If the document does not answer something{{audience}}would ask, put it in Open questions rather than guessing. - Preserve the author's claims as claims; do not endorse them, and do not fact-check beyond flagging internal contradictions. - If the text you receive is truncated, excerpted, or too long to process fully, say so at the top of the brief and state which parts you actually saw. - If the document is short enough that summarizing adds no value, say so in one sentence and give only the Gist. Tone: neutral, precise, no filler. Write for a busy professional, not a student. No phrases like "this interesting document" or "in conclusion".
Variables
Replace these placeholders with your own values before using the prompt.
{{audience}} | Who the brief is for, stated concretely (e.g. "a CFO preparing for a board meeting"). Drives what gets extracted and how points are ordered. |
|---|---|
{{priority_topic}} | The topic this reader cares about most (e.g. "pricing changes"); related points are elevated to the top of Key points. |
When to use it
- Digesting industry reports and whitepapers before a strategy meeting
- Getting up to speed on long email threads or transcripts after time away
- Triaging research papers to decide which deserve a full read
- Preparing briefing notes from lengthy policy or regulatory documents
Usage notes
Practical guidance for getting the most out of this prompt:
- Fill the audience variable specifically — "a CFO preparing for a board meeting" yields a very different brief than "a new hire catching up", and the ordering logic only works with a concrete reader.
- Paste the document's table of contents along with the text when it has one; the skip/read map gets much more accurate when section boundaries are explicit.
- The "numbers that matter" section is where hallucination risk concentrates — spot-check any figure you plan to quote against the source before forwarding the brief.
- For documents beyond the model's context window, split at section boundaries rather than mid-paragraph, then ask for per-part briefs plus a merged gist.
FAQ
What does the "Long Document Speed-Reading Summarizer" system prompt do?
Distill reports, papers, and transcripts into a layered brief: a two-sentence gist, evidence-backed key points, and open questions tuned to your audience. It belongs to the Productivity & Assistants category and is free to copy and adapt.
Which models work well with this prompt?
We recommend running it with Claude Sonnet 4.5 and Gemini 2.5 Pro and GPT-4o — chosen because the prompt's structure (length, constraints, output format) plays to their strengths. These are recommendations based on the prompt's design, not benchmark results; a formal cross-model testing program is in progress.
How do I customize this prompt?
Replace the placeholders before use: "audience" (Who the brief is for, stated concretely (e.g. "a CFO preparing for a board meeting"). Drives what gets extracted and how points are ordered.); "priority_topic" (The topic this reader cares about most (e.g. "pricing changes"); related points are elevated to the top of Key points.). Then paste the whole text as the system message of your chat or API call.