A prompt pack for academic visibility

Every draft a prompt hands you goes out under your name. The reviewer sees your name on the abstract, the public venue sees your name on the answer, the conference program prints your name next to the talk. So review every output to the last letter before you send or publish it. A good name is worth more than the time the prompt saved you. Ecclesiastes 7:1 puts it as “a good name is better than fine oil,” and Nassim Taleb calls the same idea skin in the game. The prompt has none. You have all of it.

That is the whole posture for this post. What follows is a swipe file: four prompts you can paste, fill in, and run. Each comes with a “when to use” line and a one-line caveat. Several other prompts in this field guide (the plain-language summary, the outreach email, the reconnection email, the bio, the AI-search audit) live in their own posts in this series. This post collects the remaining four.

The reviewer-readable abstract prompt

When to use: you suspect a grant reviewer outside your sub-sub-field couldn’t, in 30 seconds, place where your contribution sits. This keeps technical precision but optimizes for fast field-placement.

You are a grant-panel reviewer who is a respected researcher in a field adjacent to, but not the same as, the field of the abstract you are about to receive.

Rewrite the abstract so that *you* could understand it in 30 seconds. You may keep technical terms that are well-known one field over. You must define or replace any term that is only known inside the sub-sub-field.

Structure:
1. One sentence on the question, phrased to connect to a broader body of work the reviewer would know.
2. One sentence on the method, named by its broader family (e.g. "a regression-discontinuity design", "a transformer-based encoder", "a randomized double-blind trial") rather than its sub-sub-field name.
3. Two to three sentences on the result, including the single most important number.
4. One sentence on the implication for the broader field, not just for the sub-sub-field.

Total length: 150-200 words. No marketing language. No "novel", "innovative", or "groundbreaking".

Abstract:
<PASTE ABSTRACT HERE>

Caveat: if the rewrite feels like it loses precision, that is the point. Reviewer-readable is not specialist-readable. Both versions should exist; this one goes in the grant proposal and the executive summary.

The discussion-post prompt

When to use: you found a recurring question on a public venue (StackExchange, a subreddit, a mailing list) and want a thoughtful, non-spammy answer that links back to your work.

You are a thoughtful, slightly cynical researcher who hates self-promotional content but loves explaining things clearly.

I will give you (a) a public question, (b) the abstract or summary of my paper, and (c) which sub-question of the public question my paper answers.

Draft a response that:
- Opens by answering the public question directly. Do NOT mention my paper in the first paragraph.
- In paragraphs 2-3, gives the substantive answer using plain language. Use a number if there is a number. Cite 1-2 well-known references that are not mine.
- In the final paragraph, says: "Full disclosure, I'm one of the people who worked on this. Here is our paper, which goes into <SUB-QUESTION>: <LINK>". Then mentions one open question my paper does NOT answer, to demonstrate honesty.

Total length: 250-400 words. No marketing language. No "we're excited to share". No "groundbreaking".

(a) Public question: <PASTE>
(b) My paper summary: <PASTE PLAIN-LANGUAGE SUMMARY>
(c) Sub-question my paper answers: <PASTE>

Caveat: these posts are findable for years. Make sure the answer helps even if the reader never clicks the link. The link is the cherry, not the cake.

The conference-talk-recycling prompt

When to use: you’ve a 30-minute talk and an invitation to give it to a different audience. Adapt without rewriting.

You are helping me adapt a talk I already give well to a new audience.

I will give you:
(a) the abstract of my talk as it currently is,
(b) the new audience: their field, their seniority, what they already know, what they don't.

For the new audience, propose:
1. A revised one-paragraph abstract (~120 words) optimized for THIS audience.
2. Three slides that should be ADDED at the start to bridge from what they know to what my talk assumes.
3. One slide that should be REMOVED because it assumes background this audience lacks.
4. One sentence I should add to the closing slide that connects my work to a question THIS audience cares about.

Do NOT propose rewriting the whole talk. Reuse is the point.

(a) Current talk abstract: <PASTE>
(b) New audience: <DESCRIBE>

Caveat: the model is good at proposing bridge slides and mediocre at choosing the slide to cut. Trust yourself on the cut.

The graphical-abstract prompts (peer + public)

When to use: you’ve a plain-language summary and want a visual companion in two versions. The LLM produces a layout brief, not the image; you render it in Figma, BioRender, Canva, or an image tool. Prompt A is for a peer audience and takes your abstract. Prompt B is for the general public and takes your plain-language summary.

Prompt A, peer audience:

You are a science illustrator preparing a graphical abstract for an academic audience. I will give you the abstract of a peer-reviewed paper.

Produce a layout brief (not an image) that a designer can render. Specify:
1. The layout structure, in 3-5 sentences: how many panels, how they are arranged, what flows where.
2. The text inside each box, 10 words or less per box.
3. The single key visual anchor: the one image or diagram that carries the result.
4. One bold number to feature.
5. One sentence on how this work extends prior work in the field.

Constraints:
- Must be readable at thumbnail size.
- Technical terms are fine for this audience.
- No marketing words. No "novel", "innovative", "groundbreaking".

Abstract:
<PASTE ABSTRACT HERE>

Prompt B, general public:

You are a science illustrator preparing a graphical abstract for a general audience. I will give you a plain-language summary of a research paper.

Produce a layout brief (not an image) that a designer can render. Specify:
1. A sequential narrative layout an outsider can follow in 10 seconds: the order of panels, left to right or top to bottom.
2. A plain-language caption for each panel.
3. One striking hook for the first panel.
4. One bold takeaway for the final panel.
5. A note on color, mood, and any analogy or metaphor that makes the idea land.

Constraints:
- No jargon.
- No academic-paper aesthetics.
- The first thing the viewer sees must answer "why does this matter to me?".

Plain-language summary:
<PASTE PLAIN-LANGUAGE SUMMARY HERE>

Caveat: the LLM gives a layout brief, not the image. Render it yourself and run a thumbnail-readability test before publishing.

Read it before you sign it

Same caveat that opened this post, now as a pointer: never sign your name to a draft you’ve not read to the last letter. The prompt is a draft generator, not a sender. It produces the structure; you do the part that has your name on it.

This post is part of the Loud Camel field guide to academic visibility. You can read the whole guide as a single PDF: the handbook.