Is the bottom line written or selected?

Selected. It is the highest-ranked sentence in the document, quoted exactly, with its centrality, frequency and position scores printed underneath. If the sentence the ranking chose is not the real bottom line, edit it by hand — the tool will not invent one.

What does key-term coverage tell me?

It is the share of the document's top ten content terms, weighted by how often they appear, that the brief's sentences contain. A high coverage figure means the brief kept the vocabulary the document leans on; a low one is a signal that the selection may have missed a point that appears rarely.

Why does the bottom line sometimes come from the introduction?

Because centrality rewards sentences that share content words with the rest of the document, and an introduction that previews the topic tends to share the most. The scoring has no idea what a recommendation is, so a document that buries the point can produce a brief that buries it too.

Can I use this for a one-page brief of a long report?

Yes, with the understanding that the brief is a quotation of the report's own sentences. Long reports with repeated terminology work best; a report that never repeats a term gives the ranking no centre and the brief becomes arbitrary. Read the coverage figure before you trust the shape.

Does the document leave my browser?

No. The scoring runs in this browser, there is no model, no upload and no storage, and the brief exists only on screen and in whatever you copy or download. Closing the tab removes it.

Executive Summary Builder

Source document

Paste the full document

Brief length0 sentence(s)

The document is analysed in this browser and never uploaded. Nothing is stored after the tab closes.

Overview

Start typing above — the results update as you write.

How the brief is built

Every line above is a sentence quoted from the document. Scoring is the shared extractive method: TextRank centrality (weight 0.55) + log term frequency (0.35) + position (0.10), normalised to the best sentence. The bottom line is simply the highest-ranked sentence; the findings are the rest of the selection in document order.

This is an extraction, not an executive’s judgement. It rewards sentences that share vocabulary with the rest of the document, so an introduction can outrank a recommendation, and a document with no repeated terms has no clear centre. It does not know what a recommendation, a risk or a number is. Read every line against the source before circulating the brief, and edit by hand where the ranking missed your point.

The executive summary builder turns an extractive selection into the shape an executive reads first: one bottom line, then numbered findings in document order, with the coverage of the document's key terms reported beside them.

What is Executive Summary Builder?

An executive summary is usually written after the document by someone who has read it. This executive summary builder takes a different route and says so: it extracts. The bottom line is the highest-ranked sentence in the document, the findings are the rest of the selected sentences in the order they appear, and every line is quoted verbatim. Nothing is composed, nothing is condensed and nothing is smoothed into an executive voice.

The shape of the page is the shape of the output. A single bottom-line panel carries the highest-scoring sentence with its three score parts — centrality, frequency and position — printed underneath, so the claim that this is the most representative sentence can be checked rather than believed. The findings below are numbered in document order, each with its score, and the overview reports four figures that matter for a brief: how many sentences were selected, the document's word count, the brief's word count, and the key-term coverage, which is the share of the document's top ten content terms by weight that the selection contains.

Length is a dial rather than a guess. Short takes about 20 percent of the sentences, standard about 30 and full about 40, always rounded up to at least one sentence. The scoring is the shared extractive method used across this domain: TextRank centrality at weight 0.55, log term frequency divided by the square root of sentence length at 0.35, and a mild position decay at 0.10, normalised so the best sentence reads 1.000. Because the selection is deterministic, the same document produces the same brief on every run, in every browser.

The limits are worth reading before the brief is circulated. The ranking rewards sentences that share vocabulary with the rest of the document, which means an introduction that restates the topic can take the bottom line from a recommendation buried in the middle. Documents with one dominant term produce a brief that circles that term; short documents — under about six sentences — give the percentage dial almost nothing to work with. The scorer has no concept of a recommendation, a risk, a number or a decision: it knows term frequency and sentence connectivity, and nothing else. Headings, captions and list items are treated as sentences because the splitter is mechanical.

Used as a first pass, the tool saves the mechanical half of the work. Run it on a draft report, read the bottom line and the findings, and treat them as a shortlist of the sentences the document itself emphasises. Rewrite the bottom line in your own words if the brief is going to a reader who expects a recommendation, and check the coverage figure: a low number means the selection kept the vocabulary but may have missed a key point that appears only once.

Everything runs locally. The document is not uploaded, no model is called, nothing is stored after the tab closes, and the brief can be copied or downloaded as plain text. That combination — local, deterministic and quoted — is what makes the output auditable: any reader can find every line of the brief in the source document.

How to use Executive Summary Builder

Paste the full document; the bottom line and findings appear as you type.
Choose short, standard or full to set how much of the document the brief covers.
Read the bottom-line sentence with its score, then the numbered findings in document order.
Check the key-term coverage figure to see what the selection kept and what it left behind.
Copy or download the brief, then edit the bottom line by hand for the audience it is going to.

When to use Executive Summary Builder vs related tools

Use this builder when the output needs the bottom-line-then-findings shape. When the same selection should read as a continuous paragraph instead, the Text Summarizer presents it that way, and when the brief should be a ranked list of points rather than numbered findings, the Key Points Extractor does that.

If the material is meeting notes rather than a document, the Meeting Summary Formatter sorts decisions, actions and questions before any summary is written.

Privacy & Security

This tool runs entirely in your browser — no data ever leaves your device. There is no server round-trip, no upload, no logging, and no account required. Your input is processed locally using client-side JavaScript and is never stored, transmitted, or accessible to anyone else. When you close the tab, everything disappears.