Docsbook
Overview

List memory

MCPlist_memoryREAD$0.008 / call

THE BRIEF FOR THIS PROJECT — the owner's own goals beside the standing one (goal), what nobody here has answered yet (question), and what is already known about it (fact / rule / preference). Free on every plan. 🔴 READ THIS BEFORE YOU DECIDE ANYTHING HERE, in the same breath as get_workspace. standing_goal is every project's — be found, on Google and in AI answers — and needs no declaring; a goal line here is what this owner wants ON TOP of it. The open questions are what a previous run could not work out; answer what you can this run rather than re-deriving it. The rest is where the answers you were about to re-derive live: which page is the real pricing page, which words this product never uses, which section nobody may restructure. Each line says WHO wrote it (owner typed it in the panel, agent wrote it through this tool) and what it rests on (evidence — a URL, a path, a call id; null when there is none, which is honest for a preference and a warning sign on a fact). An owner's rule outranks your own reading of the site. A goal or a question also carries resolution — the outcome, or the answer. null there means OPEN, and the open ones are the rows worth acting on. gap names the one thing missing from this brief, when something is; it is advice about the store itself, not about the docs. 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.

kindstringoptional

Only lines of this kind. Omit for all of them. One of: fact, rule, preference, goal, question.

open_onlybooleanoptional

Only lines still open — goals not yet met, questions not yet answered. Ignored for the kinds that cannot be closed (a fact is never 'open').

Call it#

{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "list_memory",
    "arguments": {
      "kind": "fact"
    }
  }
}
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":"list_memory","arguments":{"kind":"fact"}}}'

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/list_memory
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/list_memory' \
  -H 'Authorization: Bearer dbk_YOUR_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{"args":{"kind":"fact"}}'

Updated

Was this page helpful?