AI Checker Humanizer 远程 MCP 服务器是什么?
AI Checker Humanizer 的远程 MCP 服务器是一个托管的 Model Context Protocol 端点,地址是 https://aicheckerhumanizer.com/api/mcp,AI 助手无需在本地安装或运行任何东西即可连接。

@AI Checker Humanizer
Check a draft for AI-written text and humanize it from your MCP client. The detector can return sentence scores for English and Portuguese, or a whole-text score and verdict for Spanish and French.
https://aicheckerhumanizer.com/api/mcpclaude mcp add aicheckerhumanizer --transport http https://aicheckerhumanizer.com/api/mcp --header "Authorization: Bearer ach_live_..."工具列表暂未提供
当前收录暂时没有可展示的工具列表,不代表服务没有工具。请查阅项目文档了解支持的能力。
此服务需要认证,获取工具列表可能需要登录或使用服务提供方的凭据。
点击上方按钮,直接连接服务器端点实时获取工具列表。
A hosted MCP server for AI Checker Humanizer. Your assistant can check a draft for AI-written text, rewrite it, and look up your credit balance. For a direct connection, use a Streamable HTTP client and set the Authorization header. Requests sent from web pages on other sites are rejected. If your client can launch a local stdio server instead, use the bridge described below.
https://aicheckerhumanizer.com/api/mcpAuthorization: Bearer ach_live_...Create a key on your account page. Eligible accounts get 5,000 trial credits, valid for 30 days, when they create an API key. You need a verified email that isn't a disposable address, and the trial is available once per account and once per email address.
| Tool | Arguments | Returns |
|---|---|---|
detect_ai_text | text (up to 10,000 words), optional idempotency_key | When there's a verdict, a 0–100 score and one or more labels. English and Portuguese also get per-sentence scores. Spanish and French get an overall score and verdict for the whole text. |
humanize_text | text (up to 3,000 words). Optional: mode (standard by default, or academic, casual, concise), strength (medium by default, or light, heavy), idempotency_key | The call waits up to 50 seconds for a new rewrite and returns the text if it's ready. If the rewrite is still running, pass the returned taskId as task_id to get_humanize_result. |
get_humanize_result | task_id | The task status and any rewritten text. |
get_balance | none | Credits left and when they expire. |
Retrying within 24 hours with the same idempotency_key, text and options doesn't charge you twice. Retrying a completed check returns the saved result. Retrying an accepted rewrite returns the same task ID. You can fetch the rewritten text for about 70 minutes after submission.
claude mcp add --transport http aicheckerhumanizer \
https://aicheckerhumanizer.com/api/mcp \
--header "Authorization: Bearer ach_live_..."
For Cursor, put this in ~/.cursor/mcp.json for all projects. If you use a project's .cursor/mcp.json instead, set the header to Bearer ${env:ACH_API_KEY} so the key stays out of the repository.
{
"mcpServers": {
"aicheckerhumanizer": {
"url": "https://aicheckerhumanizer.com/api/mcp",
"headers": { "Authorization": "Bearer ach_live_..." }
}
}
}
If you already use the Local harness in VS Code Chat, add this to .vscode/mcp.json and enter your API key when prompted. For a new Copilot setup, add the server under MCP Servers in Chat Customizations.
{
"inputs": [
{ "type": "promptString", "id": "ach-api-key", "description": "AI Checker Humanizer API key", "password": true }
],
"servers": {
"aicheckerhumanizer": {
"type": "http",
"url": "https://aicheckerhumanizer.com/api/mcp",
"headers": { "Authorization": "Bearer ${input:ach-api-key}" }
}
}
}
Make sure Claude Desktop can run npx from your Node.js installation. Add this to claude_desktop_config.json and restart the app. The mcp-remote bridge runs over stdio and forwards your Authorization header:
{
"mcpServers": {
"aicheckerhumanizer": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://aicheckerhumanizer.com/api/mcp", "--header", "Authorization:${ACH_AUTH}"],
"env": { "ACH_AUTH": "Bearer ach_live_..." }
}
}
}
Detection costs 1 credit per scored word when the detector returns a verdict. For each chunk that's rewritten successfully, humanizing costs 5 credits per word in the original chunk. Chunks shorter than 10 words are billed as 10. At $10 for 250,000 Starter credits, that's $0.04 per 1,000 scored words or $0.20 per 1,000 billable rewrite words. Larger packs give you more credits per dollar. Purchased credits expire after 365 days.
MCP and REST share one API key and one credit balance. Word limits, concurrency limits and idempotency work the same way on both.
AI Checker Humanizer 的远程 MCP 服务器是一个托管的 Model Context Protocol 端点,地址是 https://aicheckerhumanizer.com/api/mcp,AI 助手无需在本地安装或运行任何东西即可连接。
把端点 https://aicheckerhumanizer.com/api/mcp 添加到任意支持 MCP 的客户端,比如 Claude Code、Cursor 或 VS Code。本页的接入代码块可以一步配置好每个客户端。
需要。AI Checker Humanizer 需要一个 API key——具体获取方式请查看官方文档,拿到后填进客户端配置里即可。
AI Checker Humanizer 暴露的是 Streamable HTTP 端点,这是远程 MCP 服务器普遍使用、主流客户端都支持的传输方式。
MCP server for apimodels.app — image, video, chat and speech models with one API key. The model can review its own images and iterate.
AIsa is one key for 950+ data APIs across SEO and AI visibility, finance, social, web search, sales and agent mail — the go-to-market data stack behind prospecting, competitor research and brand monitoring. Five tools ra
Write, schedule and publish LinkedIn posts in your own voice, reply to comments and read your analytics, all from your AI assistant. Remote MCP server with OAuth sign-in.
Search and read current AWS documentation, check regional availability and load AWS agent skills, with no AWS account needed.
Find venues and submit booking inquiries through MCP.
评论