Perplexity HTML stripper

Perplexity renders every inline citation as a linked element pointing back to its source list, rather than as a plain bracketed number, so copying an answer copies that link markup along with the prose. This tool removes it, along with any other tags, attributes, script and style blocks the answer carries, and decodes a handful of common entities.

The pass runs client-side, in JavaScript, in your own browser. There is no request sent to a server to log, queue or inspect, which is also why there is no run limit: nothing here costs anything to serve, so nothing is metered.

Nothing to configure, and no step that depends on network latency.

Step 1

Paste the answer copied from Perplexity, citation numbers included.

Step 2

Run the strip. It executes instantly, on your own device.

Step 3

Read the output, checking for any unusual entity the pass left behind.

Perplexity does not run a single fixed model for every query. Depending on the selected mode, it can route a question to different underlying language models, with its own retrieval and citation layer applied consistently on top regardless of which model produced the prose. The citation and heading markup therefore stays fairly stable from answer to answer, while the underlying sentence structure can shift depending on the backend model that generated it.

This tool operates on the resulting markup rather than on any assumption about which model produced it, so the distinction has no bearing on how the stripping itself works.

A tag is a structural instruction, an angle-bracket wrapper telling a renderer how a piece of text should display or link. An entity is a character encoded as text because the literal character would be ambiguous or unsafe to include directly: & for an ampersand, < and > for the angle brackets that would otherwise be read as the start of a tag.

A citation link can carry no entities at all, while a separate portion of the same answer can carry a stray entity with no citation markup nearby. This tool performs both operations in the same pass, but they remain independent problems.

Four entities are decoded automatically:   to a normal space, & to &, and < and > to their literal angle brackets. Those four account for most of what breaks when markup-carrying text lands in a plain field.

Numeric entities and less common named entities are left exactly as written. Review the output for any stray entity code still sitting in the text as literal characters.

It is a regex-based pass built around the common case of a Perplexity answer, not a validating HTML parser. It does not reconstruct or preserve the source list a citation number pointed to, since that list lives outside the copied text entirely. It does not touch hidden Unicode characters the answer may carry alongside the citation markup, which the Perplexity zero-width space remover handles separately. There is no batch mode or API.

Researchers and analysts quoting a Perplexity answer in a report use it so citation numbers don't surface as dead links pointing at a source list that only existed on Perplexity's own page. Editorial teams pulling a Perplexity summary into a CMS or a document use it for the same reason any linked markup gets stripped before publishing.

Anyone evaluating Perplexity answers pulled from different modes uses it to normalise the formatting before comparing the underlying prose.