model.tools — Zero-Auth Agent Utilities

tools.model由 model.tools 发布官方发布者v1.3.0更新于 Oct 30, 2026

Zero-auth email OTP extractor, KV checkpoints, webhook bins, wake timers, and markdown reader.

已验证Streamable HTTP可网页运行效率与工作流网页搜索与抓取开发者工具

概览

AI 生成的概览

零认证远程工具服务器,为智能体提供一次性邮箱验证码读取、持久检查点、Webhook 收件箱、定时唤醒、任务回执与网页转 Markdown。

可以做什么
  • 读取一次性邮箱并提取最新的验证码、验证链接或魔法链接,并可布设邮箱在会话结束后继续收信
  • 保存并读回键值检查点,让多步任务在容器重置后继续
  • 向公开 Webhook 收件箱投递并轮询事件,以及安排延迟 HTTP 回调稍后唤醒智能体
  • 签发可验证的任务回执、留下只能读取一次的交接值给其他智能体,并布设看门狗在停止上报时告警指定网址
  • 向三席智能体小组提问,获得各自回答与多数票结果
  • 抓取公开网页并返回干净的 Markdown,无需浏览器
适用场景
适合运行在临时或沙箱环境中的自主智能体:需要完成邮箱验证步骤、跨重启保留状态、在 NAT 后接收 Webhook,或延迟后恢复工作。也适合需要证明任务确实执行过或想参考其他智能体意见的场景。
运行要求
远程 Streamable HTTP 端点;声明无需注册、API 密钥或 OAuth。另提供可选的 npx stdio 桥接,需要 Node.js。需要能访问服务商的网络端点。
安装前请注意
邮箱按名称开放且无需注册,知道邮箱名的人可能读到其中的邮件,不要用于敏感邮件。检查点、Webhook 收件箱和交接值同样以调用方自选的 id 为键,交接值只能读取一次。定时唤醒和看门狗会向你提供的回调网址发送 HTTP POST,请只指向自己控制的端点。任务回执与小组回答来自服务商,应视为其单方声明。

安装

在 SourceWeft 中

  1. 打开 控制台中的 model.tools — Zero-Auth Agent Utilities,将其添加到工作区。
  2. 为需要使用其工具的对话启用该服务。

Web executable,通过 Streamable HTTP。 远程服务在工作区中配置后即可从网页运行时运行。

交给智能体安装

把这段提示词粘贴到 Claude Code、Codex、Cursor 等智能体,或 SourceWeft 对话中。智能体会读取该服务器的安装指南,先向你展示发布者、端点和登录方式,经你同意后再把它添加到你的客户端。密钥不会写进任何命令。

阅读 https://sourceweft.com/mcp/tools.model%2Fmodel-tools-mcp/install.md,并按其中说明把这个 MCP 服务器接入我的客户端。修改任何配置之前,先告诉我它的发布者、端点和登录方式,等我确认。不要把 API 密钥写进命令或我会提交的文件;问我密钥应从哪里获取。

查看智能体所依据的安装指南

其他 MCP 客户端

把它添加到你客户端的 mcpServers 配置中。

{
  "mcpServers": {
    "model-tools-zero-auth-agent-utilities": {
      "type": "http",
      "url": "https://mcp.model.tools/mcp"
    }
  }
}

README

model.tools MCP (model-tools-mcp)

Zero-auth utilities for autonomous agents when a workflow needs an instant email verification code, a key-value checkpoint across container restarts, an inbound webhook bin behind NAT, a delayed HTTP wakeup callback, or clean URL-to-Markdown extraction — with no signup, no API key, and no OAuth.

  • Remote Streamable HTTP Endpoint: https://mcp.model.tools/mcp
  • REST API Twin: https://api.model.tools
  • Agent Discovery (llms.txt): https://docs.model.tools/llms.txt
  • Website: https://model.tools

Quick Connect

1. Remote Streamable HTTP (Recommended)

Add directly to Claude Desktop, Cursor, Windsurf, Cline, or any Streamable HTTP MCP client:

json
{  "mcpServers": {    "model-tools": {      "url": "https://mcp.model.tools/mcp"    }  }}

2. Stdio Bridge (npx)

json
{  "mcpServers": {    "model-tools": {      "command": "npx",      "args": ["-y", "github:maceip/model-tools-mcp"]    }  }}

Available Tools (10)

ToolDescription
get_email_verification_codeRead <inbox>@outbox.sh (or HTTP inbox <inbox>) and extract the latest 6-digit OTP code, verification link, or magic link. Any inbox name works on demand with zero registration.
save_agent_checkpointStore or read a JSON/UTF-8 checkpoint at key id (PUT/GET https://api.model.tools/api/v1/state/:id) so multi-step runs survive container resets.
schedule_agent_wakeupQueue a delayed HTTP POST callback to callback_url after delay_sec (1–86400s) to wake or resume a sleeping agent loop.
receive_webhook_eventsPost or poll JSON webhook events at hook_id (https://api.model.tools/api/v1/hook/:id) when running inside a NATted or serverless sandbox.
read_webpage_as_markdownFetch any public http(s) webpage and return clean Markdown stripped of scripts and layout boilerplate.
issue_task_receiptIssue a content-addressed task completion receipt (r_...) that downstream callers or peers can verify by URL.
arm_tripwire_inboxArm a watch on an inbox or state key that fires a JSON alert to alert_url the moment another party reads it.
ask_agent_panelSubmit a question to a 3-seat reference panel (alpha, beta, gamma) and receive individual responses plus a majority vote.
handoff_dead_dropStore a read-once secret or token (ttl_sec) for peer-agent handoff; reading once burns the drop (410 Gone thereafter).
arm_deadman_watchdogArm a deadman timer (deadline_sec) that POSTs to alert_url if your agent stops pinging /api/v1/watchdog/:id/ping.

Plain HTTP / curl Examples

Every MCP tool has a 1:1 zero-auth REST endpoint on https://api.model.tools:

bash
# 1. Poll an email inbox for the latest OTP / verification linkcurl -s https://api.model.tools/api/v1/inbox/my-agent-run/latest_otp
# 2. Save and read a checkpointcurl -s -X PUT https://api.model.tools/api/v1/state/run-42 \  -H 'Content-Type: application/json' \  -d '{"step":3,"cursor":"abc"}'curl -s https://api.model.tools/api/v1/state/run-42
# 3. Inspect or post to a webhook bincurl -s -X POST https://api.model.tools/api/v1/hook/build-7 \  -H 'Content-Type: application/json' \  -d '{"status":"ready"}'curl -s https://api.model.tools/api/v1/hook/build-7
# 4. Schedule a wakeup callbackcurl -s -X POST https://api.model.tools/api/v1/schedule \  -H 'Content-Type: application/json' \  -d '{"callbackUrl":"https://api.model.tools/api/v1/hook/wake-test","delaySec":30,"payload":{"resume":true}}'

License

MIT

来源:README.md,提交 1d2b588

工具

10

get_email_verification_code

Read an email inbox and get the verification code, OTP, signup link, or magic link from it. Use when a website emails you a 6-digit code or confirmation link and your runtime cannot receive mail. No signup, no API key, no SMTP credentials: any inbox name is live immediately.

输入参数inbox_prefix必填x_session_purpose

save_agent_checkpoint

Save your working state, scratchpad, or context summary to a durable key-value store so it survives container termination, sandbox recycle, or session reset. Read it back later with the same id. Use this to avoid losing progress when your runtime is ephemeral.

输入参数id必填body必填

schedule_agent_wakeup

Schedule a one-shot HTTP POST to a callback URL after a delay, so you can be woken up later without staying alive or polling. Use this when you need to check back on a slow job, a deploy, or a long-running request after your current turn ends.

输入参数payloaddelay_sec必填callback_url必填

receive_webhook_events

Append an event to a public webhook bin and read back everything that bin has received. Use this when you are behind NAT or a firewall and need a public address that other services can POST webhooks to.

输入参数body必填hook_id必填

issue_task_receipt

Get a verifiable, timestamped receipt proving a task ran, which a third party can check without an account. Use this when your user asks whether something actually happened and you need evidence rather than a claim.

输入参数detailsubject必填

arm_tripwire_inbox

Register a disposable email address that keeps accepting mail after your session ends, so you can read the result later from a different session. Use this when you sign up for a trial or service while your container is alive but need the follow-up email after it recycles.

输入参数name必填notettl_sec

ask_agent_panel

Ask a question and see how much independent agents agree on the answer, with disagreement reported rather than averaged away. Use this when you need to check your own reasoning against other agents before committing to an answer.

输入参数byqidanswerquestion

handoff_dead_drop

Leave a value that can be read exactly once by another agent, so a secret or capability passes between agents without appearing in any transcript. Use this when you delegate to a sub-agent or hand work to a later session.

输入参数key必填valuettl_secread_once

arm_deadman_watchdog

Ask to be alerted if you stop reporting in: fires a webhook to a supervisor URL you choose when your deadline passes without a ping. Use this when a task must not silently die and someone else needs to know if you fail.

输入参数id必填labelpayloadalert_url必填deadline_sec

read_webpage_as_markdown

Fetch a public webpage and convert it to clean readable Markdown without a browser or headless Chrome. Use this when you need to read a page's content but cannot render JavaScript or install a browser runtime.

输入参数url必填

版本历史

1
  1. v1.3.0最新Oct 30, 2026

评分

已满足适用检查项的 100%

该评分衡量此收录条目的完整程度,依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价,也不表示服务器安全或当前可用:可连通性是带日期的检查结果,评分也不包含安装量或用户评价。

  • 近 30 天内可连通必需25 分
  • 已声明认证方式必需10 分
  • 已列出工具必需15 分
  • 描述必需15 分
  • 图标10 分
  • 文档15 分
  • 许可证10 分

其他 Model 集成