How to get a WeChat Official Account's articles and full text

The job: you follow a competitor, a trade publication or a regulator on WeChat, and you want its recent articles in a form you can work with: titles, dates and reads in a table, and the complete text of the pieces that matter. Inside WeChat that means scrolling the account's history on a phone. Here it is three calls in a row: find the account, list its articles, fetch the ones you want in full. You end with a CSV of the article list and a text file per article.

No WeChat account, QR scan or cookie is involved. Each call is an HTTPS request with your key.

Step by step

Step 1: find the account

Official Account search takes query, the account's name or a keyword. Each result has id, the original ID (gh_ plus 12 characters), username (the WeChat ID, when the account has one), name, avatar_url and followers. With view=full, ext adds biz (the __biz value in the account's links), last_published_at and articles_per_week, which help you tell the real account from a lookalike.

curl -G "https://api.okdata.online/v1/wechat_mp/accounts/search" \
  -H "Authorization: Bearer $OKDATA_API_KEY" \
  --data-urlencode 'query=极客公园' \
  --data-urlencode 'view=full' \
  --data-urlencode 'fields=id,name,username,ext.last_published_at'

This is the one endpoint priced per item: $0.058 for each account returned, so at most $0.58 for a full page of 10. An exact account name usually returns one or a few matches and costs a fraction of that; a broad word fills the page. While the search runs, the full-page amount is held from your balance and the unused part is released right after. On a balance below a full page, the first page shrinks to the accounts your balance covers.

Check name before you keep an id; the right account is not always first. If you already know the gh_ ID or the WeChat ID, skip this step entirely. Finding an account's gh_ ID shows the free manual way too.

Step 2: list its articles

Official Account article history takes id, the gh_ ID from step 1 or the WeChat ID, and returns 10 articles per call, newest first, for $0.02. Each has title, url (the mp.weixin.qq.com link), published_at, the summary in text, the cover in images, and metrics.views and metrics.likes where WeChat shows them. With view=full, ext.position tells you where the article sat in that day's push: 1 is the headline.

curl -G "https://api.okdata.online/v1/wechat_mp/accounts/articles/list" \
  -H "Authorization: Bearer $OKDATA_API_KEY" \
  --data-urlencode 'id=gh_7c2a9e41b05d'

For older articles, send page.next_cursor back as cursor, unchanged, until page.has_more is false. Five pages cover about the last 50 articles.

Step 3: fetch the full text

Official Account article detail takes the article link as url and returns the body for $0.02. text is the article as plain text, paragraphs and headings on their own lines, inline images dropped; the cover stays in images. You also get title, published_at and author (the account's gh_ ID and name). With view=full, ext.author_byline is the writer named under the title, and ext.copyright and ext.source_url help you spot reposts.

curl -G "https://api.okdata.online/v1/wechat_mp/articles/get" \
  -H "Authorization: Bearer $OKDATA_API_KEY" \
  --data-urlencode 'url=https://mp.weixin.qq.com/s/Qm3xK7pVd2LrT9wYfZc1aB'

Reads and likes are null on the detail; take them from the history list.

Step 4: put it together in Python

The script finds the account by exact name, lists five pages, saves the list to CSV and fetches the three most-read articles as text files.

import csv
import os
import re
import requests

API = "https://api.okdata.online"
HEADERS = {"Authorization": f"Bearer {os.environ['OKDATA_API_KEY']}"}
NAME = "极客公园"
HISTORY_PAGES = 5  # 10 articles per page, one call each
FULL_TEXT = 3      # articles to fetch in full


def get(path, **params):
    body = requests.get(API + path, headers=HEADERS, params=params, timeout=90).json()
    if body["code"] != "ok":
        raise RuntimeError(f"{body['code']}: {body['message']} (charged: {body['charged']})")
    return body


accounts = get("/v1/wechat_mp/accounts/search", query=NAME)["data"]["items"]
account = next((a for a in accounts if a["name"] == NAME), None)
if account is None:
    raise SystemExit(f"no exact match; got {[a['name'] for a in accounts]}")

articles, cursor = [], None
for _ in range(HISTORY_PAGES):
    body = get("/v1/wechat_mp/accounts/articles/list", id=account["id"], **({"cursor": cursor} if cursor else {}))
    articles += body["data"]["items"]
    if not body["page"]["has_more"]:
        break
    cursor = body["page"]["next_cursor"]

with open("articles.csv", "w", newline="", encoding="utf-8-sig") as f:
    out = csv.writer(f)
    out.writerow(["published_at", "title", "reads", "likes", "url"])
    for a in articles:
        out.writerow([a["published_at"], a["title"], a["metrics"]["views"], a["metrics"]["likes"], a["url"]])

for a in sorted(articles, key=lambda a: a["metrics"]["views"] or 0, reverse=True)[:FULL_TEXT]:
    full = get("/v1/wechat_mp/articles/get", url=a["url"])["data"]["item"]
    with open(re.sub(r"[^\w-]", "_", full["id"]) + ".txt", "w", encoding="utf-8") as f:
        f.write(f"{full['title']}\n{full['published_at']}\n{full['url']}\n\n{full['text']}\n")

If reads are null for this account, sort by date instead and take the newest. A whole 500-article history with full text would be $11.00 after the search.

Good to know

• Reads and likes are not always there. The history list fills metrics.views and metrics.likes where WeChat shows them; otherwise they are null and meta.unavailable gives the reason. Comments, shares and saves are not available.
• Follower counts are estimates. followers in search comes from a third-party database, not from WeChat, and ext.followers_note says so.
• Deleted is free. An article taken down after it was listed returns not_found and costs nothing.
• Retries without paying twice. Send an Idempotency-Key header with each call. The same request with the same key within 24 hours returns the stored answer with meta.replayed: true and no new charge.
• Trial credit. A new key's $0.10 covers only part of a search page, so the first search may return fewer accounts than a full page. Start from a known WeChat ID, or top up first.
• Rights. The articles belong to their authors. Use them for research and monitoring, and keep the source link when you quote.

To export a whole account rather than recent articles, see how to export all articles of an Official Account. For a running watch on several accounts, see monitoring competitor Official Accounts.

{
  "endpoint": "wechat_mp.accounts.search",
  "params": {
    "query": "36氪"
  },
  "idempotency_key": "<a new UUID>"
}