Botcard
Cited AI crawler registry (training, search or user fetch) plus robots.txt checks. No fetching.
使うべきか
品質と安全性
ツール定義とプロトコルへの準拠に関する自動分析に基づいています。
コンテキストコスト
これは、サーバーのツールがモデルのコンテキストに読み込まれるたびに消費されるおおよそのトークン数です。数が多いほど、ほかのタスクに使える注意が減ります。
インストール
ワンクリックインストール
これを `claude_desktop_config.json` ファイルに追加してください:
{
"mcpServers": {
"botcard": {
"url": "https://whosthatbot.com/mcp"
}
}
}リモートエンドポイント
https://whosthatbot.com/mcpstreamable-httpできること
ツール一覧
ツール（4）
🟢check_robots(robots_txt, path, ids)
Match a robots.txt file the caller pastes against Botcard's AI crawler tokens, using RFC 9309. Each row is allow or disallow plus that token's vendor stance. This tool does not fetch any website. Do not send url, host, target, website, robots_url, or fetch.
入力スキーマ
{
"type": "object",
"properties": {
"robots_txt": {
"type": "string",
"description": "The robots.txt file contents. Not a URL."
},
"path": {
"type": "string",
"description": "Path to test, starting with /. Defaults to /."
},
"ids": {
"type": "array",
"items": {
"type": "string"
},
"description": "Card ids to check, such as gptbot. Omit to check every card."
}
},
"required": [
"robots_txt"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}🟢list_crawlers
List the AI crawler robots.txt tokens in this dataset. Each row names the vendor page the summary was checked against. This tool does not fetch any website. Do not send url, host, target, website, robots_url, or fetch.
入力スキーマ
{
"type": "object",
"properties": {},
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}🟢get_crawler(id)
Return one card by id, including the vendor source URL and the date it was read. This tool does not fetch any website. Do not send url, host, target, website, robots_url, or fetch.
入力スキーマ
{
"type": "object",
"properties": {
"id": {
"type": "string",
"description": "Card id, such as gptbot or google-extended."
}
},
"required": [
"id"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}🟢suggest_robots(allow, block, policy, path)
Write a robots.txt from this dataset for the kinds you allow or block. Each group cites the vendor page. effect request_only means the vendor does not say the crawler obeys that line. control_token is not training. The file has no User-agent: * group. This tool does not fetch any website. Do not send url, host, target, website, robots_url, or fetch.
入力スキーマ
{
"type": "object",
"properties": {
"allow": {
"type": "array",
"items": {
"type": "string",
"enum": [
"training",
"search",
"user_fetch",
"ads_check",
"control_token",
"owner_requested",
"open_crawl",
"safety",
"preview",
"other"
]
},
"description": "Kinds to Allow, such as search."
},
"block": {
"type": "array",
"items": {
"type": "string",
"enum": [
"training",
"search",
"user_fetch",
"ads_check",
"control_token",
"owner_requested",
"open_crawl",
"safety",
"preview",
"other"
]
},
"description": "Kinds to Disallow, such as training."
},
"policy": {
"type": "string",
"description": "Same choice in words, such as \"allow search, block training\"."
},
"path": {
"type": "string",
"description": "Path the Allow or Disallow names. Defaults to /."
}
},
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}コミュニティ
エビデンス