AI crawler

cohere-training-data-crawler

Cohere's crawler, which downloads web content used as training data for its language models.

Operator: Cohere
Category: AI crawler
Label: Unknown; identified by its self-description only

Allow or block cohere-training-data-crawler

cohere-training-data-crawler is identified by its self-description only, so it is reported as a bad bot. This rule denies it by name. Switch the action to allow if you trust it.

{
  "name": "Deny cohere-training-data-crawler",
  "enabled": true,
  "match": "all",
  "conditions": [
    {
      "field": "botId",
      "op": "eq",
      "value": "cohere-training-data-crawler"
    }
  ],
  "action": "deny"
}

Add it in the console rule editor, or start from the rule templates.

Check a request from your server

Send the request your server received to the bot check with your secret key. The answer says whether the visitor is a good bot, a bad bot or not detected, and which entry it matched, for example cohere-training-data-crawler.

curl -X POST https://api.fpchecker.com/v1/bot-check \
  -H 'X-Secret-Key: sk_...' \
  -H 'Content-Type: application/json' \
  -d '{"method":"GET","url":"https://yoursite.com/pricing","ip":"203.0.113.7","headers":{"user-agent":"..."}}'

Read the integration guide

Some entries draw on ai.robots.txt (MIT licence). ai.robots.txt on GitHub