TextClean MCP
Free deterministic PDF, Markdown, citation, invisible-character, and formatting cleanup for agents.
我該用這個嗎
品質與安全性
B
根據工具定義與協定合規性的自動化分析。
上下文成本
~167Token(工具定義)
~1.5 KB典型回應大小
極小的注意力影響(128k 上下文的 0.13%)
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": {
"textclean": {
"url": "https://textclean.org/mcp/free"
}
}
}遠端端點
https://textclean.org/mcp/freestreamable-http它能做什麼
工具清單
工具(1)
🟢 唯讀🟡 寫入🔴 刪除⚪ 未知
⚪clean_text(text, destination, preset, stripMarkdown, removeCitations, ...)
Clean copied text while preserving headings, paragraphs, lists, quotations, emphasis, and intentional spacing. Supports PDF repair and destination-specific rendering.
輸入結構描述
{
"type": "object",
"properties": {
"text": {
"type": "string",
"minLength": 1,
"maxLength": 20000
},
"destination": {
"type": "string",
"enum": [
"preserve",
"email",
"message",
"blog",
"document",
"plain"
]
},
"preset": {
"type": "string",
"enum": [
"default",
"pdf",
"ai",
"html",
"list"
]
},
"stripMarkdown": {
"type": "boolean"
},
"removeCitations": {
"type": "boolean"
},
"replaceLongDashes": {
"type": "string",
"enum": [
"preserve",
"hyphen",
"comma",
"colon",
"space"
]
}
},
"required": [
"text"
],
"$schema": "https://json-schema.org/draft/2020-12/schema"
}社群
證據
近期觀測
已驗證未記錄版本1 個工具