Summarize a document
Turn a long document into a faithful summary: main point first, key points by importance, then decisions and open questions.
Task: Summarize a document · Other tasks
Fill in the details
Paste the whole document, not a description of it. Plain text works best; headings and lists can stay as they are.
Optional. Naming the reader helps Claude choose the level of detail and which terms to explain.
Optional. Topics you care about most.
Copy your prompt
This site doesn't run Claude or show model output. Results depend on your input and the model you use.
You've edited the prompt, so changes to the fields aren't applied.
What you type is kept in this tab's session storage so a reload doesn't lose it. Use "Clear this task" to remove it. Browsers can restore session data when they reopen tabs, so closing a tab isn't a guaranteed way to erase it.
When to use this template
- You have a report, article, transcript or policy that is too long to read closely, and you need its main point and the facts that matter.
- You need to brief someone else, such as a manager or a new teammate, and want the summary pitched at their level.
- You want a summary you can check against the source, because the prompt asks for points in order of importance and for decisions and open questions in a separate list.
When not to use it
- You need exact quotes or a legal reading of the text. A summary paraphrases by design; ask for quotations with locations instead.
- The material is many separate documents that disagree with each other. Summarize them one at a time, or use a comparison prompt that asks Claude to list where they differ.
- You want Claude to add outside knowledge or opinions. This template deliberately limits Claude to what the document says.
Why this structure
- The document comes first, inside
<document>tags, and the instructions follow it. Anthropic's long-context guidance recommends placing long material above the question, and the tags help Claude tell your text apart from the instructions. - The three-part structure (main point, key points by importance, then decisions and open questions) spells out what a good answer looks like, so the prompt asks for the same shape each time you use it.
- The length options describe the result in words and units, such as "about 200 words" or "3 to 5 bullet points", instead of vague terms like "brief".
- The optional reader line only appears when you fill it in, so the prompt stays short when you do not need it.
- The last paragraph asks Claude to stay within the document and to flag unclear or missing parts. That gives Claude explicit permission to say "this is not covered" instead of filling the gap.
Example input (fictional)
- Document
Quarterly update from the Harbor Street community garden (fictional). Plot fees stay at 40 per season. The rain barrel project is two months late because the supplier changed. Volunteers asked for a second tool shed; the committee will vote in May. Water use dropped 12 percent after the drip lines were installed.
- Length
Short (3 to 5 bullet points)
- Who will read the summary
new volunteers
Follow-ups to send Claude
- Which of these points are stated directly in the document, and which did you infer? Quote the sentence for each.
- Rewrite the summary for someone who has five minutes and no background on this topic.
- List every number and date in the document with the sentence it comes from.
Common mistakes
- Pasting a link instead of the text. Claude may not be able to open it, and you cannot check what it read. Paste the content itself.
- Asking for a "short" summary without saying how short. Pick a length option so the result is predictable.
- Trusting a summary of a document that was cut off. If your paste was truncated, the summary will quietly miss the end; the last instruction asks Claude to point this out, but check the end of your paste too.
Related templates
- Turn meeting notes into action items: Turn rough meeting notes into a clear list of action items with owners and due dates, plus decisions made and questions left open.
- Extract data to JSON: Pull specific fields out of messy text into JSON that matches your field list, with clear rules for missing or ambiguous values.
- Edit a draft for clarity: Make a draft clearer for a named reader while keeping your meaning and voice, with a short list of what changed and why.
Sources
- Prompting best practices: Long context prompting (Anthropic documentation)
- Prompting best practices: Structure prompts with XML tags (Anthropic documentation)
- Reduce hallucinations (Anthropic documentation)