How to Collect Cerfia public social data via API

Quick Answer

Cerfia fact-checking workflows can pull claim-related posts with TwexAPI search operators, then enrich the dataset with reply threads and promotion flags. Store source URLs, retrieval times, raw JSON, and any 100-ID batch boundaries so every annotation remains auditable.

FAQ

Why use TwexAPI instead of the first-party developer API for this workflow?

For fact-check pipelines, TwexAPI provides Bearer-authenticated workflows documented at https://docs.twexapi.io. A typical post or profile read uses 5 Credits (about $0.05 per 1,000 under the published conversion), and qualifying paid plans publish 20+ QPS. New accounts receive 20,000 starting Credits. As of 2026-08-20, first-party developer API Post reads list at $5 per 1,000 resources and User reads at $10 per 1,000; official rate limits vary by endpoint and access level.

How much does this API workflow cost on TwexAPI?

A typical read uses about 5 Credits. Under the published conversion, 1,000 such reads use 5,000 Credits, or about $0.05; a 10,000-read job uses about 50,000 Credits. Actual costs vary by endpoint, so confirm the endpoint price and current plan at https://twexapi.io/pricing.

When you need data from a public X account, keep the workflow concrete: collect the profile snapshot, page through recent posts, and run topic-specific searches only when you need a filtered subset.

This guide uses CerfiaFR as the example screen name. If the account handle changes, replace the screen name before running the requests.

Collection Plan

Answer: Collection Plan means using TwexAPI Bearer APIs on api.twexapi.io for this user case. A typical read uses 5 Credits (about $0.05 per 1,000), and qualifying paid plans publish 20+ QPS. As of 2026-08-20, first-party developer API Post and User reads list at $5 and $10 per 1,000 resources; rate limits vary by endpoint.

Use three TwexAPI endpoints depending on the data you need:

Task	Endpoint	When to use it
Get public account metadata	`GET /twitter/{screen_name}/about`	Capture profile fields before collecting posts.
Page through the account timeline	`POST /twitter/{screen_name}/timeline/page`	Collect recent posts with cursor pagination.
Search within public X results	`POST /twitter/advanced_search/page`	Filter by query, such as `from:CerfiaFR climate`, and paginate search results.

Store both the raw response and the normalized rows. The raw response gives you an audit trail if your parser changes later.

Get Account Metadata

Answer: Get Account Metadata means using TwexAPI Bearer APIs on api.twexapi.io for this user case. A typical read uses 5 Credits (about $0.05 per 1,000), and qualifying paid plans publish 20+ QPS. As of 2026-08-20, first-party developer API Post and User reads list at $5 and $10 per 1,000 resources; rate limits vary by endpoint.

curl --request GET \
  --url https://api.twexapi.io/twitter/CerfiaFR/about \
  --header 'Authorization: Bearer <token>'

Keep this profile snapshot separate from the post table. It can include account identifiers, display fields, description text, and public counters returned by the API.

Page Through Timeline Posts

Answer: Page Through Timeline Posts means using TwexAPI Bearer APIs on api.twexapi.io for this user case. A typical read uses 5 Credits (about $0.05 per 1,000), and qualifying paid plans publish 20+ QPS. As of 2026-08-20, first-party developer API Post and User reads list at $5 and $10 per 1,000 resources; rate limits vary by endpoint.

POST /twitter/{screen_name}/timeline/page accepts count and next_cursor. Use the first response's next_cursor for the next page while has_next_page is true.

curl --request POST \
  --url https://api.twexapi.io/twitter/CerfiaFR/timeline/page \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
    "count": 20
  }'

The response includes the current data page plus pagination fields.

{
  "code": 200,
  "msg": "success",
  "data": [],
  "has_next_page": true,
  "next_cursor": "DAAHCgAB...",
  "requested_count": 20,
  "effective_count": 20,
  "parsed_count": 20
}

Python Collector

Answer: Python Collector means using TwexAPI Bearer APIs on api.twexapi.io for this user case. A typical read uses 5 Credits (about $0.05 per 1,000), and qualifying paid plans publish 20+ QPS. As of 2026-08-20, first-party developer API Post and User reads list at $5 and $10 per 1,000 resources; rate limits vary by endpoint.

This collector retrieves a bounded number of timeline pages and produces review-friendly rows.

1import os
2from typing import Any
3
4import requests
5
6TOKEN = os.environ["TWEXAPI_BEARER_TOKEN"]
7SCREEN_NAME = "CerfiaFR"
8BASE_URL = "https://api.twexapi.io"
9
10def headers() -> dict[str, str]:
11    return {"Authorization": f"Bearer {TOKEN}", "Content-Type": "application/json"}
12
13def get_profile(screen_name: str) -> dict[str, Any] | None:
14    response = requests.get(
15        f"{BASE_URL}/twitter/{screen_name}/about",
16        headers=headers(),
17        timeout=30,
18    )
19    response.raise_for_status()
20    return response.json().get("data")
21
22def get_timeline_page(
23    screen_name: str,
24    *,
25    count: int = 20,
26    next_cursor: str | None = None,
27) -> dict[str, Any]:
28    payload: dict[str, Any] = {"count": count}
29    if next_cursor:
30        payload["next_cursor"] = next_cursor
31
32    response = requests.post(
33        f"{BASE_URL}/twitter/{screen_name}/timeline/page",
34        headers=headers(),
35        json=payload,
36        timeout=30,
37    )
38    response.raise_for_status()
39    return response.json()
40
41def normalize_post(post: dict[str, Any]) -> dict[str, Any]:
42    tweet_id = post.get("tweet_id") or post.get("id")
43    return {
44        "tweet_id": tweet_id,
45        "created_at": post.get("created_at_datetime") or post.get("created_at"),
46        "text": post.get("full_text") or post.get("text") or "",
47        "favorite_count": post.get("favorite_count") or 0,
48        "retweet_count": post.get("retweet_count") or 0,
49        "reply_count": post.get("reply_count") or 0,
50        "quote_count": post.get("quote_count") or 0,
51        "url": f"https://x.com/i/status/{tweet_id}",
52    }
53
54def collect_timeline(screen_name: str, max_pages: int = 3) -> list[dict[str, Any]]:
55    rows: list[dict[str, Any]] = []
56    cursor = None
57
58    for _ in range(max_pages):
59        page = get_timeline_page(screen_name, count=20, next_cursor=cursor)
60        rows.extend(normalize_post(post) for post in page.get("data", []))
61
62        if not page.get("has_next_page"):
63            break
64        cursor = page.get("next_cursor")
65        if not cursor:
66            break
67
68    return rows
69
70if __name__ == "__main__":
71    profile = get_profile(SCREEN_NAME)
72    posts = collect_timeline(SCREEN_NAME, max_pages=2)
73    print(profile.get("screen_name") if profile else SCREEN_NAME, len(posts))

Optional Topic Search

Answer: Optional Topic Search means using TwexAPI Bearer APIs on api.twexapi.io for this user case. A typical read uses 5 Credits (about $0.05 per 1,000), and qualifying paid plans publish 20+ QPS. As of 2026-08-20, first-party developer API Post and User reads list at $5 and $10 per 1,000 resources; rate limits vary by endpoint.

Use timeline collection when you want account posts. Use advanced search by page when you need a topic-filtered subset.

curl --request POST \
  --url https://api.twexapi.io/twitter/advanced_search/page \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
    "searchTerms": ["from:CerfiaFR climate"],
    "sortBy": "Latest"
  }'

Save the query with each collected row. Without that, it is hard to explain later why a post appeared in a dataset.

Data Hygiene

Answer: Data Hygiene means using TwexAPI Bearer APIs on api.twexapi.io for this user case. A typical read uses 5 Credits (about $0.05 per 1,000), and qualifying paid plans publish 20+ QPS. As of 2026-08-20, first-party developer API Post and User reads list at $5 and $10 per 1,000 resources; rate limits vary by endpoint.

Record screen_name, collection time, endpoint, query, and cursor for every run.
Deduplicate by tweet_id.
Store raw JSON before normalizing fields.
Do not treat interaction counts as proof of accuracy or importance.
Review media, quoted posts, and replies before using the data in reports.

Wrap-up

Answer: Wrap-up means using TwexAPI Bearer APIs on api.twexapi.io for this user case. A typical read uses 5 Credits (about $0.05 per 1,000), and qualifying paid plans publish 20+ QPS. As of 2026-08-20, first-party developer API Post and User reads list at $5 and $10 per 1,000 resources; rate limits vary by endpoint.

For a public account such as CerfiaFR, the practical flow is: capture the profile, page through the timeline, optionally run topic-specific searches, and normalize the results into reviewable rows.

That gives you useful account data without adding unsupported claims or noisy analysis.