Library
Every prompt in the collection
Search by what you want to do, or narrow by task, the kind of model you use and how much adapting you want to do. Filters are kept in the page address, so you can bookmark a view.
Search and filters need JavaScript. Every prompt is listed below, and each task has its own page:
Prompts
11 of 11 prompts
-
PL-01
Summarise a document and point to the passage behind every claim
A summary you can check: each point quotes the sentence it came from, and anything the document does not say is marked as missing.
Understand · Starter · 4 placeholders
Works with:- General chat assistant, good fit
- Model API, good fit
- Small local model, workable with limits
-
PL-02
Turn rough meeting notes into decisions, actions and open questions
Sorts messy notes into what was decided, who does what by when, and what is still open, without inventing owners or dates.
Understand · Starter · 3 placeholders
Works with:- General chat assistant, good fit
- Small local model, good fit
- Model API, workable with limits
-
PL-03
Rewrite a draft for a different reader without changing the facts
Adjusts vocabulary, order and length for a new audience, then lists every change of meaning it was tempted to make so you can check none slipped in.
Write · Starter · 4 placeholders
Works with:- General chat assistant, good fit
- Model API, good fit
- Small local model, workable with limits
-
PL-04
Get a critical review of a draft before you send it
Asks the model to read as your actual reader and find what is unclear, missing or likely to be misread, instead of polishing sentences.
Write · Practitioner · 4 placeholders
Works with:- General chat assistant, good fit
- Reasoning model, good fit
- Small local model, workable with limits
-
PL-05
Learn a concept by being asked questions, not lectured
Sets the model up as a patient tutor that finds out what you know, explains one step at a time and checks your understanding before moving on.
Learn · Starter · 3 placeholders
Works with:- General chat assistant, good fit
- Reasoning model, workable with limits
- Small local model, workable with limits
-
PL-06
Review a code change for bugs, risks and missing tests
A structured review of a diff: correctness first, then security and edge cases, then tests, with a confidence level on every finding.
Code · Advanced · 3 placeholders
Works with:- Reasoning model, good fit
- General chat assistant, good fit
- Model API, good fit
- Small local model, workable with limits
-
PL-07
Explain an error message and what to check first
Turns a stack trace or error into a plain explanation, the most likely causes in order, and one quick check for each.
Code · Practitioner · 3 placeholders
Works with:- General chat assistant, good fit
- Reasoning model, good fit
- Small local model, good fit
-
PL-08
Compare options against criteria you set in advance
Builds a comparison table from your criteria and constraints, separates facts you gave from assumptions it made, and says what would change the answer.
Plan · Practitioner · 4 placeholders
Works with:- Reasoning model, good fit
- General chat assistant, good fit
- Small local model, workable with limits
-
PL-09
Run a pre-mortem: imagine the plan failed and find out why
Assumes the plan has already failed, lists the most plausible reasons, and turns each into an early warning sign and a cheap precaution.
Plan · Practitioner · 2 placeholders
Works with:- Reasoning model, good fit
- General chat assistant, good fit
- Small local model, workable with limits
-
PL-10
Extract fields from text into a table, marking what is missing
Pulls the fields you name out of emails, forms or notes into one table, and writes missing instead of guessing a value.
Analyse · Practitioner · 3 placeholders
Works with:- General chat assistant, good fit
- Model API, good fit
- Small local model, good fit
-
PL-11
Group open-ended feedback into themes with counts and quotes
Sorts survey comments or reviews into themes, counts how many responses mention each, and quotes real examples so the themes can be checked.
Analyse · Practitioner · 4 placeholders
Works with:- General chat assistant, good fit
- Reasoning model, good fit
- Small local model, workable with limits
No prompt matches
Nothing matches the current search. The collection is small on purpose: remove a filter, or try a broader word such as summary, review or plan.