Short version
We publish compiled cryptocurrency market data. Compiling it costs us money, so large-scale automated collection of the rate pages is licensed. Everything else stays open.
Free — no licence, no key
Search and assistant agents that cite the source and send readers back to us (OAI-SearchBot, Claude-SearchBot, PerplexityBot, ChatGPT-User, Claude-User, Perplexity-User and similar) are not limited. They bring us readers; that is a fair exchange.
Conventional search engines (Google, Bing, Yandex and others) are not limited.
Editorial content — news and articles — is open to every agent.
Everyone else gets a free allowance of 100 rate-page requests per day, so any agent can see and evaluate the data.
Licensed — bulk collection of rate pages
Beyond the free allowance, automated collection of market-rate pages by training and
dataset crawlers (GPTBot, ClaudeBot, CCBot, meta-externalagent, Bytespider, Amazonbot,
Applebot-Extended, Google-Extended and similar) requires a licence. Those requests are
answered with HTTP 402 Payment Required and a link to this page.
This is not a block. It is a price.
How to get access
Preferred: the JSON API. Structured, versioned, documented, cheaper for both sides than scraping HTML — see /api/docs.php. The free tier needs no key (500 requests/day); paid tiers start at 25,000 requests/day.
Bulk or training licence: write to [email protected] with the agent name, the volume you need and what the data will be used for. We answer commercial requests.
Technical notes
Gated responses carry X-AI-License: required, Retry-After and a
Link: rel="license" header pointing here. Classification is by User-Agent:
agents that identify themselves honestly are treated according to the rules above.
Agents that disguise themselves as browsers are not covered by any of these allowances.
Our robots.txt states the same policy in machine-readable form.