Get project doc outline
/api/v1/get_project_doc_outlineList every markdown page in the workspace with a short summary (title, heading count, char count) plus its lifecycle — status, version and agentMayBuildFrom. An INVENTORY, not a search: NOT the way to answer a question or find the page about something — that is search_project_docs. Building this list downloads and parses every page of the site (slow on a large one), and a list of titles answers nothing by itself. It is also the REVIEW BOARD for documentation that is governed by status: status: 'review' lists what is waiting on a human, status: 'generated' what a machine wrote that nobody has read. counts always describes the whole prefix, not the filtered rows, so 'four of ninety pages are approved' is one call. Statuses: generated (A machine wrote this page and no human has read it yet.) · draft (Someone is still writing it. Not ready to be read as settled.) · review (Waiting for a human to read it and decide.) · approved (A human read this version and signed off. Safe to build work from.) · locked (Frozen on purpose. Agents may read it and build from it, but may not rewrite it.) · deprecated (Superseded. Kept so its links keep working, not to be relied on.) · archived (History. Neither built from nor edited.)
Price — $0.00003 per call (twice what serving it costs us), charged to the workspace balance, the same as over MCP.
Also reachable by name at POST /api/v1/tools/get_project_doc_outline.
Authorization: Bearer dbk_YOUR_API_KEY.Sent from your browser straight to the API — never to Docsbook, never stored.counts still covers them all.200 with ok: false: the call was made and billed, and that refusal is its answer.- Use it for 'what does this documentation cover', to pick a folder for a NEW page, or after
search_project_docsfound nothing.
ok to see whether it succeeded.curl 'https://docsbook.io/api/v1/get_project_doc_outline' \
-H 'Authorization: Bearer dbk_YOUR_API_KEY'{
"ok": true,
"result": "<result>",
"duration_ms": 0
}