Every change Wellknown observed on this MCP server, newest first, with what it was before and what it became. Tool-surface changes carry the definition diff. Nothing here is edited after the fact.
Changed the definition of "account" and "search"
⟨12 unchanged words⟩ used today and left, rate limit, resultlimits.limits, extraction points left for today (clusters.extract_points_left). Spends nothing. In clusters, max is how ⟨14 unchanged words⟩
⟨62 unchanged words⟩ quote an exact phrase ("angular.min.js"); inside quotes punctuation only separates words: "jquery-ui.min.js" finds the words jquery, ui, min, js in that order; several words must all appear; a leading minus ⟨265 unchanged words⟩
Changed the definition of "search"
⟨39 unchanged words⟩ its ready-made query on the examples pages (read_docs,:e.g.theexamples/chatTechnologies pages listed in its table of contents), not by its name. Query syntax: ⟨279 unchanged words⟩
Changed the definition of "account", "extract_from_cluster", "read_docs" and 1 more
The account behind the token: the current time (now, UTC), plan, searches used today and left, rate ⟨24 unchanged words⟩
⟨27 unchanged words⟩ name ready-made presets - email (email addressesanywhere in the page source (image names, Sentry and example addressesfromleftmailto:out)links), phone (phone numbers from tel: links
Authorization required → not required, issuer https://publicwww.com
Changed the definition of "extract_from_cluster"
⟨27 unchanged words⟩ name ready-made presets - email (email addressesfromanywheremailto:inlinksthe page source (image names, Sentry and example addresses left out)), phone (phone numbers from tel: links ⟨234 unchanged words⟩
Changed the definition of "build_signatures", "combine_clusters" and "search"
⟨23 unchanged words⟩ result is a ready search - pass itsquery tosearch (thesearch toolquery in double quotes) to the search tool to find more such websites. Best on a cluster ⟨8 unchanged words⟩ a mixed cluster may give none. Sorted bestfirst:first; take the one with the highest in_cluster - how many of the websites looked at have the code (coverage: "4 of 5", in_all: every one of them), websites - rough count in the whole ⟨27 unchanged words⟩
Changed the definition of "create_cluster" and "search"
⟨23 unchanged words⟩ "With query: take only the first Nresultswebsites, in the search's own order (popularity, ⟨8 unchanged words⟩ up to the plan's limit. With depth:it countsseveral pagesbefore they fold intoofwebsites,onesowebsitethemakeclusteronecandomain,hold
Changed the definition of "create_cluster", "extract_from_cluster", "get_cluster" and 1 more
⟨91 unchanged words⟩ in the index, no cluster is created. 'dropped' in the answer counts what was left out: not_a_domain, duplicate, not_in_index, over_limit. The plan caps the size of a cluster. ⟨18 unchanged words⟩
⟨90 unchanged words⟩ company profile links), youtube (youtube.com channel links and video links (youtu.be, embed)), tiktok (tiktok.com profile links), pinterest ⟨170 unchanged words⟩
Changed the definition of "extract_from_cluster" and "search"
⟨143 unchanged words⟩ slashes, the first capture group is the value (with no group, the whole match), e.g. /data-site-id="([0-9]+)"/i. Up ⟨109 unchanged words⟩
⟨65 unchanged words⟩ , site:example.com lists the pages of onesite;site, -site:com or -site:example.com leaves them out; depth:all searches inner pages too (by default ⟨99 unchanged words⟩ place by popularity, 1 being the mostpopular.
Changed the definition of "extract_from_cluster"
⟨216 unchanged words⟩ a website, not only the home page. Every call is paid for, a repeat of the same websites too: keep the answer and go on from next_offset. The REST API gives the same as a ⟨15 unchanged words⟩
First tool surface recorded: 11 tools (server version 1.3.0)
Authorization required, Bearer, issuer https://publicwww.com
Authorization required, Bearer, issuer https://publicwww.com
Authorization required, Bearer, issuer https://publicwww.com
Authorization required, Bearer, issuer https://publicwww.com
Authorization required, Bearer, issuer https://publicwww.com
Certificate changed, valid to 2027-01-11
Authorization required, Bearer, issuer https://publicwww.com
Authorization required, Bearer, issuer https://publicwww.com
Certificate recorded, valid to 2026-11-13
Authorization required, Bearer, issuer https://publicwww.com
Unknown → Live
https://api.publicwww.com/mcp (mcp_streamable_http) — from mcp_registry, with the record
Showing the latest 23 events. The API returns up to 500 and filters by kind: ?kind=tool_surface_changed
⟨6 unchanged words⟩ , in Markdown: query syntax and operators, websites about a topic (like:), regular expressions (snipexp:), country domains, ready-made ⟨64 unchanged words⟩
⟨34 unchanged words⟩ :"string"},"page":{"description":"Page id from the table of contents, for example query-syntax/regex, api/limitsLeaveoroutprices.pageLeaveandoutpages to get the table of contents. Otherwise a page id from it.","type":"string"},"pages":{ ⟨26 unchanged words⟩
⟨134 unchanged words⟩ without depth: there is no sites).SimilarWebsiteswebsitesabout a topic or like a given one, most similar first instead of by popularity: ⟨177 unchanged words⟩
⟨13 unchanged words⟩ - in any of them, 'diff' -inthe first clusterbut not inminus the second (exactly two, order matters).: clusters [7, 3] keeps the domains of 7 that are not in 3. Spends nothing.
⟨26 unchanged words⟩ code fragment. Results are ranked by site popularity. Look a technology up by its ready-made query on the examples pages (read_docs, e.g. examples/chat), not by its name. Query syntax: quote an exact phrase ("angular.min.js ⟨154 unchanged words⟩ ); with words, like: orders their results. like: is the topic of a site, not its code: websites that share code with given ones are found by build_signatures on a cluster of them. With like:, total counts everything the filters ⟨73 unchanged words⟩
⟨241 unchanged words⟩ against the account's daily allowance. At most 10requestscalls a minuteacrossofallsearch and the cluster tools,andshared with the RESTAPI:API;onaccount and read_docs do not count. On too_many_requests wait retry_after seconds. Ask for a small number of rows first.
⟨67 unchanged words⟩ e.g. [\"email\", \"phone\"]. Presets and regex together: up to 10.","items":{"enum":["email", ⟨17 unchanged words⟩ ,"app_store"],"type":"string"},"maxItems":10,"type":"array"},"regex":{"description" ⟨13 unchanged words⟩ .","items":{"type":"string"},"maxItems":10,"type":"array"}},"required":[ ⟨3 unchanged words⟩
⟨12 unchanged words⟩ integer"},"page":{"default":1,"description":"Page number, from 1; total_pages in the answer says how many there are.","minimum":1,"type":"integer"},"rows ⟨19 unchanged words⟩
⟨107 unchanged words⟩ them (a lower bound when sites_exact isfalsefalse; without depth: there is no sites). Similar websites, most similar first instead ⟨32 unchanged words⟩ - websites about the same thing as thatone;one (it compares home pages: when a shop or landing page has too little text there, the search finds nothing - describe the topic in words instead); with words, like: orders their results. With ⟨35 unchanged words⟩ read_docs. Searches count against the account's daily allowance. At most 10 requests a minute across all tools and the REST API: on too_many_requests wait retry_after seconds. Ask for a small number of rows first.