KARON WEB ACCESS API

The API that lets AI agents
reliably read the public web

Retrieve public web pages, clean their content into markdown, text, or HTML, and use browser fallback and structured extraction when needed. The current public contract is grounded in production code and live smoke results.

# Basic content extraction for AI agents curl -sS -X POST "https://api.karonlabs.net/v1/agent/browse" \ -H "Authorization: Bearer YOUR_API_KEY" \ -H "Content-Type: application/json" \ -d '{"url":"https://docs.python.org/3/","extract":"markdown"}'

Get a Free API Key Developer Docs

Available in Production Today

Browse

POST /v1/agent/browse

Returns text, markdown, HTML, pruned, or JSON output that AI can read directly. This is the primary endpoint.

Fetch

POST /v1/fetch

Use it when you need raw HTML, status codes, headers, or cookie information. It supports sessions and caching.

Health

GET /health

Checks service status for operational monitoring and deployment smoke tests.

Watch

/v1/watch/*

Stores snapshots of public pages and calculates diffs against previous results for document change monitoring.

JSON

extract: "json"

Extracts the structure you need from page content. You can use `json_schema` and `json_prompt` together.

MCP

Agent Client Integration

Use the REST API as a tool in MCP-compatible environments such as Claude Code, Cursor, and Codex.

Examples

curl -sS -X POST "https://api.karonlabs.net/v1/agent/browse" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"url":"https://docs.python.org/3/","extract":"markdown"}'

curl -sS -X POST "https://api.karonlabs.net/v1/agent/browse" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "url":"https://example.com",
    "extract":"json",
    "json_prompt":"Extract the document title and key changes.",
    "json_schema":{"type":"object","properties":{"title":{"type":"string"},"changes":{"type":"array","items":{"type":"string"}}}}
  }'

curl -sS -X POST "https://api.karonlabs.net/v1/fetch" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"url":"https://www.w3.org/TR/"}'

curl -sS -X POST "https://api.karonlabs.net/v1/watch/snapshot" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"url":"https://example.com","extract":"text","ttl_days":1}'

Credit Accounting

cache/curl 1

A cached response or successful lightweight fetch or extraction costs 1 credit.

browser 10

A successful browser path, including `full_access`, `wait_selector`, or automatic fallback, costs 10 credits.

structured extraction +5

A successful structured extraction with `extract: "json"` adds 5 credits to the base cost.

Use cost_credits in the response as the authoritative amount charged. Current credit and rate-limit details are also available from GET /v1/pricing.

Add Public Web Access to Your Stack

Turn official documentation, standards, research papers, and technical references into formats AI agents can read.

Get a Free API Key View Docs

The API that lets AI agentsreliably read the public web