ClapHereBot
ClapHereBot fetches individual web pages on the explicit request of a ClapHere customer. It is a user-directed agent, not a crawler.
What it does
ClapHere is a business platform for restaurants and food businesses. Its assistant lets an operator ask questions like "what's on this supplier's price list" or "what changed on this menu page" about a page they name. When that happens, ClapHereBot retrieves that single page and extracts its readable text so the assistant can answer.
What it does not do
| Crawl | It fetches only pages a customer explicitly names or selects from search results. It does not follow links, traverse sitemaps, or discover URLs on its own. |
| Train models | Retrieved content is used to answer the request that triggered it. It is not used as training data for any model. |
| Republish | Content is not stored as a public copy, indexed, or served to anyone other than the customer who requested it. |
| Bypass access controls | It does not solve CAPTCHAs, use residential proxy networks, or disguise its identity as a browser. If a site declines the request, the request fails and the customer is told so. |
How to identify it
Two independent signals. Either is sufficient.
1. Cryptographic signature (preferred)
Every page request is signed with Web Bot Auth
(RFC 9421 HTTP Message Signatures, Ed25519). Requests carry
Signature-Agent, Signature-Input and Signature headers.
Our public keys are published at:
https://claphere.com/.well-known/http-message-signatures-directoryA signature that verifies against a key in that directory is genuinely from ClapHere. A request claiming to be ClapHereBot without a valid signature is not from us.
2. User-Agent
Mozilla/5.0 (compatible; ClapHereBot/1.0; +https://claphere.com/bot)A User-Agent string alone is trivially forged by anyone. Use the signature if you need certainty.
Request volume
Low and bursty by nature: one request per page, triggered by a person asking a question. There is no scheduled or bulk fetching. If ClapHereBot is generating more load on your site than you expect, that is a bug on our side and we want to hear about it.
If you would rather we did not
Email bot@claphere.com with the domain and we will add it to a permanent do-not-fetch list. Blocking the User-Agent or the signature key at your edge also works and we will not attempt to work around it.