POST/api/v1/me/analytics/ai/ask

Ask your analytics (AI)

Answers a question about your own analytics with Claude Haiku 5.5, which picks from a fixed set of queries that Markdrop runs on the documents you own (totals, top links, a link's numbers, referrers, channels, devices, countries, views by day, period comparisons, file versions). Streams server-sent events: status (what it's checking), delta (answer text; reset: true clears what was shown), then done with answer and links (only analytics pages the queries returned); or limit, or error. The same question the same day is answered from the copy kept (cached: true) unless fresh is true. Counts towards your AI limits.

API token required20 per minute per IPMCP: ask_analytics

Guide: Insights and Ask

Code samples and Try it

Request body

application/json, required

questionstringRequired
1–500 characters.
rangestring
One of 7d, 30d, 90d, all. One of 7d, 30d, 90d, all.
freshboolean
Default false.

Responses

200OK.
401No token, or the token is revoked or mistyped.
422The request didn't validate. detail lists each problem, with where it is (loc) and why (msg).
429Too many requests from your IP address. Wait the seconds in Retry-After.

curl -X POST https://api.markdrop.in/api/v1/me/analytics/ai/ask \ -H "Authorization: Bearer $MARKDROP_TOKEN" \ -H "Content-Type: application/json" \ -d '{"question":"Which link did best this month?","range":"30d"}'

const res = await fetch("https://api.markdrop.in/api/v1/me/analytics/ai/ask", { method: "POST", headers: { Authorization: `Bearer ${process.env.MARKDROP_TOKEN}`, "Content-Type": "application/json", }, body: JSON.stringify({ "question": "Which link did best this month?", "range": "30d" }), }); const data = await res.text();

import os import requests r = requests.post( "https://api.markdrop.in/api/v1/me/analytics/ai/ask", headers={"Authorization": f"Bearer {os.environ['MARKDROP_TOKEN']}"}, json={ "question": "Which link did best this month?", "range": "30d", }, ) r.raise_for_status() print(r.text)