ForkFree

Prompts / Research and analysis

Weekly Newsletter Digest With Skip List

You get a short digest that groups the week's newsletters into themes, picks out what is genuinely new, and lists what you can ignore. It works because the AI must compare each item with what you have already seen and justify every skip.

Worked in our testThe reply has the three required sections, 2–4 themes with issues named, about 270 words and no invented facts, and it flags the unclear items. It falls short in two ways. Several "New this week" reasons run to more than one sentence, which breaks the one-sentence rule. It also guesses that the warehouse software is "likely too big" for a solo shop.

The prompt

Act as a research editor who briefs busy readers on what actually matters in their inbox.

Task: Read the newsletter issues below and produce a weekly digest for {reader_profile}. Leave out anything already covered in {already_covered}.

Newsletter issues:
{newsletter_issues}

Steps:
1. Read every issue and list each distinct announcement, claim or link.
2. Group related items into 2 to 4 themes. Name each theme in a short phrase and say which issues mention it.
3. Pick the items that are new compared with {already_covered} and useful to this reader. Give a one-sentence reason for each.
4. Build a skip list of items that are repeated, purely promotional, off-topic or already covered, with a few words on why.
5. Do not invent details. If an item is unclear, say so rather than guessing.

Output: Three headed sections called "Themes", "New this week" and "Skip list". Use short bullets, keep it under 300 words, and write in plain British English.

Change the parts in curly braces. Use it once a week when several newsletters pile up and you need a quick, repeat-free summary.

What happened when we ran it

We sent

Act as a research editor who briefs busy readers on what actually matters in their inbox.

Task: Read the newsletter issues below and produce a weekly digest for A solo owner of a small online shop selling handmade candles. She has no technical background, spends about ten minutes a day on email, and wants practical ideas for marketing, pricing and tools that save time.. Leave out anything already covered in Last week's digest covered Instagram's new shop tab, a guide to Black Friday discount timing, and a review of a free invoicing app called TidyBooks..

Newsletter issues:
Issue A (Retail Weekly): Royal Mail has raised parcel prices by 7 percent from January. Several shops are moving to flat-rate shipping to absorb the change. There is also a sponsored link to a warehouse software trial.

Issue B (Small Shop Notes): Email open rates for independent shops averaged 38 percent this quarter. Shops that sent a short gift guide in early November saw the best results. A reader asks whether flat-rate shipping hurts sales.

Issue C (Maker Daily): A new AI tool writes product descriptions from a photo. The author tried it on 20 items and edited about half. Instagram's new shop tab is mentioned again, with a short how-to.

Steps:
1. Read every issue and list each distinct announcement, claim or link.
2. Group related items into 2 to 4 themes. Name each theme in a short phrase and say which issues mention it.
3. Pick the items that are new compared with Last week's digest covered Instagram's new shop tab, a guide to Black Friday discount timing, and a review of a free invoicing app called TidyBooks. and useful to this reader. Give a one-sentence reason for each.
4. Build a skip list of items that are repeated, purely promotional, off-topic or already covered, with a few words on why.
5. Do not invent details. If an item is unclear, say so rather than guessing.

Output: Three headed sections called "Themes", "New this week" and "Skip list". Use short bullets, keep it under 300 words, and write in plain British English.

The model replied

## Themes - **Shipping costs** (Issues A, B): Royal Mail parcel prices rise 7 percent from January. Some shops are moving to flat-rate shipping. A reader in Issue B asks whether that hurts sales, but the issue's answer isn't given. - **Email marketing** (Issue B): Independent shops averaged a 38 percent open rate this quarter. A short gift guide sent in early November performed best. - **Time-saving tools** (Issues A, C): An AI tool writes product descriptions from a photo. Issue A also carries a warehouse software trial, but it is sponsored. ## New this week - **Royal Mail 7 percent rise from January:** It affects your postage costs, so you may need to adjust prices or shipping charges soon. - **Flat-rate shipping as a response:** It is a practical option to consider, though the evidence on its effect on sales is unclear. - **Early-November gift guide email:** It is timely and low effort, and it did best for similar shops. Note that "early November" is now. - **38 percent open rate benchmark:** It gives you a figure to compare your own emails against. - **AI product description tool:** It could save writing time. The author edited about half of 20 items, so expect to check the results. The tool's name, price and quality details are not given. ## Skip list - **Instagram shop tab how-to (Issue C):** Already covered last week. - **Warehouse software trial (Issue A):** Sponsored, and likely too big for a solo candle shop. - **Black Friday timing and TidyBooks:** Not in this week's issues, and covered last week anyway.

Run on Claude Sonnet 5.5, 01 Nov 2026. Automatic check: 4 out of 5.

Mentioned for

Claude, ChatGPT, Gemini

Where it came from

Found on Web, shared by Smart Prompts for AI (Substack): Saturday Mailbag: 5 New Questions from the Survey.

This is our own version of the idea, written and tested by us. The original may differ.

More research and analysis prompts

Research and analysis Worked in our test
Blind Side-by-Side Judge for Two Versions

You get a fair, evidence-based verdict on which of two drafts, prompts or outputs is better for a stated purpose. It works because the judge sets criteria first, ignores surface polish, and checks its own result for order bias.

Found on GitHub, shared by UnderleveledBuilder
Research and analysis Struggled in our test
Goal-First Document Review With Set Answer Shape

You get a focused review of a long document in exactly the format you asked for. It works because the material comes first, the goal and answer shape come last, and the model is told what you want rather than how to think.

Found on Web, shared by Emma Al (AI Portalen), Substack, 24 Oct 2026
Research and analysis Worked in our test
Fact-Check an AI Answer Against Sources

You get a claim-by-claim audit of an AI answer, showing exactly which statements are backed by your source text and which were made up. It works because the model must quote evidence for every claim and is barred from using outside knowledge.

Found on GitHub, shared by comet-ml