Built for data products and AI agents

Social data infrastructure

Whale social media data infrastructure

Search public content, versions, metrics, and structured facts across platforms through one stable, normalized, and authorized API built for data products and AI applications.

One cross-platform schemaEntitlement-aware searchTraceable versions and metrics
POST /v1/contents/search200 OK
curl https://whale.xinzhiaigc.com/v1/contents/search \
  -H "Authorization: Bearer $WHALE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "mode": "hybrid",
    "query": "生成式 AI",
    "source_platforms": ["weibo", "youtube"],
    "top_k": 10
  }'
03datasets
25+platforms
03search modes

One contract, many sources

Turn fragmented platform data into usable content assets

Whale normalizes identity, content, transcripts, metrics, snapshots, and platform extensions. Consumers stay independent from collection differences and storage infrastructure.

Normalized content

Articles, videos, posts, products, and future types share a stable core while Parts and Facts preserve platform-specific structure.

Hybrid retrieval

Keyword, vector, and hybrid modes share one response contract with dataset, platform, language, and time filters.

Change history

Content changes create versions while time-series engagement metrics create snapshots for historical analysis.

Unified Data API

One request searches authorized content across platforms

Application keys are bound to organizations, subscriptions, and access policies. Dataset, platform, field, rate, and concurrency rules are enforced centrally.

Explore the public API

Consumer API Key

Bearer whale_consumer_...

Stable contract

keyword · vector · hybrid

Policy enforcement

dataset · platform · field · rate

Platform coverage

Coverage across major content and discovery platforms

Coverage spans Chinese and global social platforms, discovery sources, and vertical datasets. Platform-specific structure is normalized without flattening away useful details.

Chinese social

DouyinKuaishouBilibiliWeChat ChannelsXiaohongshuXianyuWeiboWeChatWeChat Official AccountZhihu

Global social

XRedditMediumLinkedInInstagramFacebookThreadsTelegramTikTokYouTube

Discovery

Google SearchGoogle NewsGoogle TrendsNewsRSSWebsiteApp StorePodcastDiscordGitHub

Vertical sources

ShengcaiChuhai Mini BookPM Radar

MCP Integration

Connect AI agents directly to Whale data

The official MCP server exposes unified search, document details, and content retrieval as tools. It reuses Consumer Key entitlements and rate limits without creating a parallel authorization path.

Read the MCP guide
01

Works with MCP hosts such as Claude Desktop and Codex

02

Local stdio process with no new public port

03

Results retain source, time, document ID, and relevance evidence

Access control

Authorization, traceability, and stable contracts by design

Organizations own applications. The platform assigns subscription and access ceilings, then applications receive keys within that boundary. Every query is authorized, rate limited, and observable.

Application API keys

Dataset and platform grants

Field-level access

Usage and error visibility

Start building your data product

Register an organization, create an application, and use the API or MCP after a subscription is assigned.