AI-powered web content analysis: summary, key points, entities, sentiment, and credibility signals for any URL.
What it does
The `/content/analyze` endpoint on content.hugen.tokyo is part of the Content Intelligence API, a suite of x402-paid endpoints for extracting, summarizing, and analyzing web content. This specific endpoint accepts a URL and returns a structured AI-generated analysis including a summary, key points, named entities (with types), sentiment label and score, topic tags, content type classification, and credibility signals (author presence, date, sources, word count adequacy).
The API uses a dual-engine extraction system that handles JavaScript-rendered pages, so you don't need your own scraping infrastructure or LLM setup. The endpoint is a simple GET request with a required `url` query parameter (max 4096 chars) and an optional `strip_query` boolean to remove query parameters before processing. Payment is $0.02 USDC per call on Base (x402 protocol). Alternative access methods include an MCP gateway, API keys with free trial credit, and a Python SDK (`x402-pay`).
The API is live and returns a 402 payment challenge with a sample response demonstrating the output structure. The OpenAPI spec (v3.1.0) is published at `/openapi.json`, and interactive docs are available at `/docs`. The service is part of a broader ecosystem hosted at hugen.tokyo that includes discovery, brokering, and intel services.
Capabilities
Use cases
- —Analyze news articles for sentiment, entities, and credibility before aggregation
- —Classify and tag web content by topic for automated content curation pipelines
- —Extract key points from long-form articles for agent-driven research workflows
- —Assess credibility signals of web sources as part of fact-checking automation
- —Build structured knowledge bases from unstructured web pages
Fit
Best for
- —AI agents that need structured analysis of arbitrary web URLs without managing scraping or LLM infrastructure
- —Content curation and monitoring pipelines requiring sentiment and topic classification
- —Research automation workflows that need entity extraction and key-point summarization
Not for
- —Bulk scraping of thousands of pages per minute (per-call pricing at $0.02 adds up)
- —Extracting raw HTML or full page content (use /content/extract instead)
- —Offline or non-URL-based text analysis (requires a live URL as input)
Quick start
# Using the x402-pay Python SDK
pip install x402-pay
import x402_pay
result = x402_pay.get(
'https://content.hugen.tokyo/content/analyze?url=https://example.com/article'
)
print(result.json())Example
Response
{
"topics": [
"artificial-intelligence",
"machine-learning"
],
"summary": "The article discusses recent advances in AI technology...",
"entities": [
{
"name": "OpenAI",
"type": "org"
}
],
"sentiment": {
"label": "positive",
"score": 0.7
},
"key_points": [
"AI models are getting smaller and faster"
],
"content_type": "news",
"credibility_signals": {
"has_date": true,
"has_author": true,
"has_sources": true,
"word_count_adequate": true
}
}Endpoint
Quality
The endpoint is live (402 challenge captured), has a published OpenAPI 3.1 spec, clear pricing ($0.02 USDC on Base), and a sample response structure from the 402 challenge. However, the response schema is not formally defined in the OpenAPI spec (empty schema objects), and there are no detailed docs beyond the auto-generated Swagger page.
Warnings
- —Response schemas in the OpenAPI spec are empty — the example response structure is inferred from the 402 challenge sample only.
- —Root URL returns 404; no dedicated documentation pages found beyond /docs (Swagger UI).
Citations
- —API described as 'Extract, summarize, and analyze web content in one call. Dual-engine extraction handles JS-rendered pages.'https://content.hugen.tokyo/docs
- —Payment is $0.02 USDC on Base per callhttps://content.hugen.tokyo/content/analyze
- —OpenAPI spec version 3.1.0 published at /openapi.jsonhttps://content.hugen.tokyo/docs
- —Alternative access via MCP gateway at https://mcp.hugen.tokyo/mcp, API keys, and x402-pay Python SDKhttps://content.hugen.tokyo/content/analyze
- —Endpoint accepts url (required, max 4096 chars) and strip_query (optional boolean) query parametershttps://content.hugen.tokyo/openapi.json