China SnapshotDocs
Get a key ↗
Open in AI ↗
Open in ChatGPTOpen in ClaudeView Markdown
Getting started

The retained record, over HTTP.

Search Chinese publications, follow public attention and add economic context—with retained history and traceable sources.

What you can build#

China Snapshot is a searchable archive and API for research, strategy and policy teams—and their AI agents. Retrieve retained records without operating your own collectors. A collected corpus is not a complete register of everything published.

Three datasets#

DatasetUnit of analysisStart here
Documents and publicationsOne retained record per source URL; current extraction and source metadataFind records
Public attentionOne sampled hot-list appearance; repeated observations over timeTrack attention
Economic indicatorsA collection observation, or one release per series and periodUse indicators

Make your first request#

No key is needed for the public datasets. Start with one retained document, then add filters or a key.

curl
curl --fail-with-body --silent --show-error --max-time 30 "https://china-snapshot.com/api/datasets?dataset=documents&limit=1"
python
import json
from urllib.request import Request, urlopen
request = Request('https://china-snapshot.com/api/datasets?dataset=documents&limit=1')
with urlopen(request, timeout=30) as response:
    data = json.load(response)
print(json.dumps(data, ensure_ascii=False))
typescript
const response = await fetch("https://china-snapshot.com/api/datasets?dataset=documents&limit=1");
if (!response.ok) throw new Error(`HTTP ${response.status}`);
console.log(await response.json());

Onboard your agent#

Give your coding agent the setup prompt. It checks the live contract, makes an anonymous request and explains key setup when needed. MCP research tools require an eligible credential; a Free key is for the public data API.

Choose your access#

Use REST for application code, CSV for analysis, MCP for entitled research clients, or the browser to explore. Plans and limits come directly from the tier configuration.

Know what is retained#

Publication time, collection time and reference period are different clocks. Documents expose current extraction, not immutable body revisions. Platform heat is not public opinion. Read the data contract before comparing periods.

Search documentation