Douyin, WeChat and Xiaohongshu data in Claude Code and Codex (2026)
Connect DataXAPI's MCP server: paste one sentence with your key into Claude Code or Codex and it sets itself up; then ask in plain words, with 230 endpoints on one key. A page of Douyin video search is $0.0200, a page of an Official Account's articles $0.0200. The free trial covers up to 333 calls; a failed call is not charged.
Paste this into Claude Code or Codex: “Read https://dataxapi.com/agents.md and set up DataXAPI in this agent. My API key is dx_…; use it for every DataXAPI call.”
Or run npx dataxapi setup in a terminal: it writes the entry and installs the base skill. It does not write your key: put that in the environment variable DATAXAPI_API_KEY.
Client
Where the entry goes
Claude Code
~/.claude.json
Codex
~/.codex/config.toml
What to ask
Say what you want; it finds the endpoint and its price, then calls it.
You say
What it calls
Search Douyin for 露营 videos from the last week and write the 20 most liked to a CSV
douyin.videos.search
The titles and reads of this Official Account's last 30 articles, as a table
Write a script that saves these 10 Douyin accounts' new videos to a database every day
douyin.users.videos.list
Why does it not fill the context?
A coding agent's context is what it can least spare. what an AI reads through DataXAPI takes 95% fewer tokens than the platform's raw JSON: a page of 7 Douyin results is 1,622 tokens, against 187k raw. Every platform uses the same field names, so the parsing code it writes works on the next platform too.
How do I keep spending in hand?
Finding an endpoint and reading its parameters and price are free; only call is charged, and a failed call is not. Give the agent its own key with a limit; in a script, X-Max-Price caps what one call may cost, and the same Idempotency-Key within 24 hours is never charged twice.
Why DataXAPI
Steady: endpoints were available 99.9% of the last 30 days, on a public status page you can check.
Fast: a paid call took 2.5 seconds at the median over the last 7 days (counted 2026-11-06).
Fewer tokens: across 226 endpoints, one recorded real response each, the platform's own JSON comes to 9.2M tokens and what an AI reads through DataXAPI's MCP server to 452k, 95% fewer; the JSON of a direct API call is 90% fewer too (counted 2026-11-06). Every platform answers in one shape: the author in author, counts in metrics, the next page in page.next_cursor, so another platform needs no new parser.
See before you pay: every endpoint's page shows a recorded real response, field by field.
Spending you control: a failed call is not charged; each key can have a limit, X-Max-Price caps one call, and the same Idempotency-Key within 24 hours is never charged twice.
Ready for AI: a remote MCP server listed in the official MCP registry as com.dataxapi/mcp, usable from Claude, Cursor and ChatGPT.
Free trial of up to 333 calls, no card.
Endpoints
Each with its price; every page has the parameters, the fields and a recorded response.