Docsbook
Overview

Get page diff impact

MCPget_page_diff_impactREAD$0.04 / call

Did a documentation edit actually work? (BUSINESS) Give it a commit and it compares the pages that commit TOUCHED against the pages it did NOT, in the week before and the week after — dead-end rate, self-serve resolution, time to first value, visit outcomes. The untouched pages are the control group, and they are the point: docs traffic moves for reasons that have nothing to do with you, so 'dead ends fell after my edit' is only evidence if they fell FURTHER on the edited pages than everywhere else. A change that merely matched the site trend is reported as no effect, not as a win. Call it with NO sha to list the commits that can be measured here — this tool carries its own index of them, so there is no separate history tool to call first. Use it after making a change you were told to make, and before making the same kind of change again. It measures a change that arrived as a COMMIT. For a change that did not — a setting, a language, a nav, a prompt — take a reading before and after with any read tool and compare the two with compare_tool_calls, which needs no commit at all. Also breaks the visits down by country, reader language and device, each beside the same slice's move on the untouched pages — which is how 'traffic went up' becomes a decision ('the growth is German readers on the English page') instead of a headline. Returns an explicit 'cannot measure' for commits too recent to have an after-window, or too old for visit data to still cover their before-window — never zeroes that look like a collapse. This answers WHAT, not what to do about it. If you have not already got the method from docsbook_expert, get it first: it names which readings answer this question, what to compare them against, and what would make the conclusion wrong. One call, changes nothing.

Arguments
workspace_idstringoptional

Workspace ID (optional when MCP endpoint is auto-scoped). Numeric workspace id — OR the project as the user names it: 'owner/repo', the repo name alone, the site's display name, its docs URL or custom domain. Text is resolved server-side; an ambiguous name returns the candidates instead of guessing, so pass what the user said rather than calling list_workspaces first.

shastringoptional

The commit to measure, full or short SHA. Omit to get the list of commits that CAN be measured here, newest first.

window_daysintegeroptional

Days compared on each side (default 7)

Call it#

{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_page_diff_impact",
    "arguments": {}
  }
}
curl -X POST 'https://docsbook.io/api/mcp/server' \
  -H 'Authorization: Bearer dbk_YOUR_API_KEY' \
  -H 'Content-Type: application/json' \
  -H 'Accept: application/json, text/event-stream' \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"get_page_diff_impact","arguments":{}}}'

Try it over REST#

The same tool is callable as a plain HTTP request, no MCP client required. It runs on the same server, at the same price.

Your workspace is resolved from the API key, so workspace_id is decided server-side here and anything you send for it is ignored.

POST/api/v1/tools/get_page_diff_impact
Authorization
Authorizationstringrequired

Your API key, sent as Authorization: Bearer dbk_YOUR_API_KEY.Sent as the Authorization header. Your key is used only by your browser for this request — it is never sent to Docsbook or stored.

Body
argsobject

The arguments above, as one JSON object.

Request
curl -X POST 'https://docsbook.io/api/v1/tools/get_page_diff_impact' \
  -H 'Authorization: Bearer dbk_YOUR_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{"args":{}}'

Updated

Was this page helpful?