rattlesnakesbymail
Dated, sourced claims about AI crawlers and agents, offered as MCP tools.
我該用這個嗎
品質與安全性
根據工具定義與協定合規性的自動化分析。
上下文成本
這是每次將伺服器的工具載入模型上下文時所消耗的約略 token 數量。數量越高,可用於其他工作的注意力就越少。
安裝
一鍵安裝
將以下內容加入你的 `claude_desktop_config.json` 檔案:
{
"mcpServers": {
"rattlesnakesbymail": {
"url": "https://rattlesnakesbymail.com/mcp"
}
}
}遠端端點
https://rattlesnakesbymail.com/mcpstreamable-http它能做什麼
工具清單
工具(5)
🟢lookup_crawler(name_or_ua)
Look up a documented crawler on Rattlesnakes By Mail by name, slug, or user-agent string, and return identity fields, current claims, publication dates, and evidence URLs. Use lookup_crawler when the crawler is already named, or when the user agent is already known to belong to a documented crawler.
輸入結構描述
{
"type": "object",
"properties": {
"name_or_ua": {
"type": "string",
"description": "A crawler name, slug, or full user-agent string."
}
},
"required": [
"name_or_ua"
]
}🟢identify_user_agent(user_agent)
Match an unrecognized raw User-Agent header against the crawlers documented on Rattlesnakes By Mail, and return the matched crawler plus instructions for verifying the request. Use identify_user_agent when the crawler behind the User-Agent header is not yet known.
輸入結構描述
{
"type": "object",
"properties": {
"user_agent": {
"type": "string",
"description": "A raw User-Agent header string."
}
},
"required": [
"user_agent"
]
}🟢list_changes(since, entity, limit)
List recorded changes to claims on Rattlesnakes By Mail, newest first, with the change date and the affected crawler. Filter by date with since, by crawler with entity, or by both. Use list_changes to detect crawler policy updates published after a given date.
輸入結構描述
{
"type": "object",
"properties": {
"since": {
"type": "string",
"description": "An ISO date; only changes on or after this date are returned."
},
"entity": {
"type": "string",
"description": "An entity slug to filter changes to one crawler."
},
"limit": {
"type": "integer",
"description": "Maximum rows to return, up to 200.",
"maximum": 200
}
}
}🟡ask(question)
Match a free-text question against the claims published on Rattlesnakes By Mail, and return the matching claims, or no match when no published claim covers the question. Every question sent to ask is recorded, and a recorded question may be published on the public questions page at Rattlesnakes By Mail. Do not send private or identifying text to ask.
輸入結構描述
{
"type": "object",
"properties": {
"question": {
"type": "string",
"description": "A free-text question about crawlers, agents, or claims."
}
},
"required": [
"question"
]
}🟡submit_note(target_type, target_id, body, token)
Submit a correction or note about a claim, an entity, or a question on Rattlesnakes By Mail. Requires a client token issued out of band by the publisher. Every note is stored as pending, and an editor reviews every pending note before publication.
輸入結構描述
{
"type": "object",
"properties": {
"target_type": {
"type": "string",
"enum": [
"claim",
"entity",
"question"
],
"description": "The kind of record the note is about."
},
"target_id": {
"type": "string",
"description": "The id or slug of the claim or entity, or the id or hash of the question, the note is about."
},
"body": {
"type": "string",
"description": "The note text, up to 2000 characters.",
"maxLength": 2000
},
"token": {
"type": "string",
"description": "A client token issued by Rattlesnakes By Mail."
}
},
"required": [
"target_type",
"target_id",
"body",
"token"
]
}建議的提示詞
list_changeslist_changes社群
證據