{
  "markdown": "# SendSoon AI MCP\n\n[English](README.md) | [简体中文](README.zh-CN.md) | [日本語](README.ja.md)\n\nSendSoon MCP is a [Model Context Protocol (MCP)](https://modelcontextprotocol.io/) server that lets Codex and Claude use [SendSoon](https://sendsoonai.com/) capabilities: send email, look up public IPs, and convert local documents to Markdown.\n\nConfigure it once, then describe tasks in natural language. You usually do not need to mention SendSoon in every prompt.\n\n## What you can do\n\n| Tool | Purpose | Example |\n| --- | --- | --- |\n| `send_email` | Send a single email | “Send me a test email” |\n| `ip_lookup` | Look up public IP information | “Look up the location of 8.8.8.8” |\n| `markitdown_convert` | Convert a local document to Markdown | “Convert this PDF to Markdown” |\n\n## Everyday examples\n\n```text\nSend an email to user@example.com with the subject “Meeting reminder” and the body “The meeting starts at 3 PM today.”\n```\n\n```text\nLook up the location and ISP for 8.8.8.8.\n```\n\n```text\nConvert /path/to/report.pdf to Markdown and summarize the key points.\n```\n\nIf the agent does not select a tool automatically, say:\n\n```text\nUse the sendsoon MCP to complete this task.\n```\n\n## Install in your AI client\n\nThe instructions below cover **Codex** and **Claude Desktop**.\n\n### Codex\n\n**Config file:** user-level `~/.codex/config.toml` (macOS / Linux) or `%USERPROFILE%\\.codex\\config.toml` (Windows).\n\n```toml\n[mcp_servers.sendsoon]\ncommand = \"npx\"\nargs = [\"-y\", \"@sendsoon/mcp\"]\n\n[mcp_servers.sendsoon.env]\nSENDSOON_API_KEY = \"\"\n```\n\n**Verify:** Save the file, restart Codex, and run `/mcp` in a chat. You should see `sendsoon` connected. In the desktop app, also check `Settings > MCP servers`.\n\n**First use:** Start a new conversation and ask for a simple task (e.g. “Look up 8.8.8.8”). Approve the tool if Codex asks for permission.\n\n### Claude Desktop\n\n**Config file:**\n\n| OS | Path |\n| --- | --- |\n| macOS | `~/Library/Application Support/Claude/claude_desktop_config.json` |\n| Windows | `%APPDATA%\\Claude\\claude_desktop_config.json` |\n\nYou can also open `Settings > Developer > Edit Config` inside Claude Desktop.\n\n```json\n{\n  \"mcpServers\": {\n    \"sendsoon\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@sendsoon/mcp\"],\n      \"env\": {\n        \"SENDSOON_API_KEY\": \"\"\n      }\n    }\n  }\n}\n```\n\n**Verify:** Save the file, **fully quit** Claude Desktop (including the system tray), then relaunch. Start a new chat and confirm `sendsoon` tools appear in the tools menu or settings.\n\n**First use:** Claude may ask to approve tool access on the first invocation; allow it to continue.\n\n## Prepare the configuration\n\nOptional environment variable:\n\n| Setting | Required | What to enter |\n| --- | --- | --- |\n| `SENDSOON_API_KEY` | No | Leave empty for an unregistered trial. One public IP can send up to three free test emails per day. Enter your generated Key for continued use |\n\nNever commit a real Key to Git or share it with anyone.\n\n### Get an API Key\n\n1. Sign up or sign in on the [SendSoon registration page](https://sendsoonai.com/login-register).\n2. Open [Profile](https://sendsoonai.com/profile) and generate a Key in the API Key section.\n3. Copy the one-time `ssk_live_...` Key immediately and enter it as `SENDSOON_API_KEY` in your MCP configuration.\n4. Save the configuration and restart your MCP client. Requests with a valid Key do not use the anonymous IP daily quota.\n\nIf the anonymous quota is exhausted, `send_email` will tell you to register and configure a Key. An invalid or revoked Key is rejected and does not fall back to the anonymous quota.\n\n## Try it in Google Colab\n\nSkip local setup and try `ip_lookup`, `markitdown_convert`, and `send_email` in the browser.\n\n[![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sendsoon/mcp/blob/main/docs/SendSoon.ipynb)\n\nThe notebook calls the same SendSoon HTTP APIs used by these MCP tools.\n\n---\n\n## Visual local testing (MCP Inspector)\n\nTo **see all three tools in a browser and click through them manually** before installing into an AI client, use the [MCP Inspector](https://github.com/modelcontextprotocol/inspector).\n\nRequires Node.js 20+. Run the command below in a terminal; Inspector opens a web UI automatically:\n\n```bash\nnpx @modelcontextprotocol/inspector npx -y @sendsoon/mcp\n```\n\n### How to test\n\n1. Wait for `MCP Inspector Web is up and running` in the terminal. If the browser does not open, paste the `http://127.0.0.1:6274?...` URL from the terminal.\n2. Confirm the page header shows **Connected**.\n3. Open the **Tools** tab in the top-right area of the page.\n4. Select a tool from the list on the left, fill in the parameters, and click **Execute Tool**.\n5. Read the response in the **Results** panel on the right. On failure, inspect `error.code` and `error.message`.\n\n## License\n\nSee [LICENSE](LICENSE).\n",
  "bytes": 4859,
  "sha": "c6daaefd0c9cb2c8c6ef3120c736b879af11e828e067fdc0271d67c3674fff47",
  "repo_slug": "sendsoon/mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_sendsoon_mcp_74dcf514/readme"
}