by nitrosend
Optional stdio bridge to Nitrosend’s remote MCP server for clients without native HTTP transport.
# Add to your Claude Code skills
git clone https://github.com/nitrosend/mcpLast scanned: 10/27/2026
{
"issues": [],
"status": "PASSED",
"scannedAt": "2026-10-27T09:38:24.386Z",
"npmAuditRan": true,
"pipAuditRan": true,
"promptInjectionRan": true
}See how mcp compares with popular alternatives.
mcp is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by nitrosend. Optional stdio bridge to Nitrosend’s remote MCP server for clients without native HTTP transport. It has 160 GitHub stars.
Yes. mcp passed SkillsLLM's automated security scan — a dependency vulnerability audit plus prompt-injection heuristics — with no high-severity issues. You can read the full report in the Security Report section on this page.
Clone the repository with "git clone https://github.com/nitrosend/mcp" and add it to your Claude Code skills directory (see the Installation section above).
mcp is primarily written in JavaScript. It is open-source under nitrosend on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh mcp against similar tools.
No comments yet. Be the first to share your thoughts!
30 days in the Featured rail · terms & refunds
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM is an informational directory and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
Automated security scan — not a certification.
Optional stdio bridge for Nitrosend. Most clients should connect directly to Nitrosend's remote MCP server at https://api.nitrosend.com/mcp; this package exists for stdio-only clients and local development.
All plans, including free plans, have full access to the generally available Nitrosend MCP/API/CLI surface. Plan limits still apply to send volume, seats, AI action allowances, and paid add-ons. The Free plan includes 8,000 free emails up front, then 500 free emails every month.
For Claude Desktop, Claude.ai, and Claude Cowork, use the remote MCP connector
UI. Do not install @nitrosend/cli or this stdio bridge for the standard
Claude connector setup.
Manage contacts, compose emails, build automated flows, and launch campaigns through natural language.
For image media used in templates or campaigns, MCP should orchestrate rather
than carry large bytes. Small chat-attached images can use nitro_ingest
with kind=image and image_data; public images can use image_url when
Nitro-hosted permanence is desired; larger local files should call
nitro_ingest(upload: {kind: "image", filename, content_type, byte_size, checksum}),
PUT bytes to direct_upload.url, then pass the returned signed_id to
nitro_ingest.
Most clients support OAuth — sign in via your browser, no API keys needed.
claude mcp add --transport http nitrosend https://api.nitrosend.com/mcp
Or add to .mcp.json:
{
"mcpServers": {
"nitrosend": {
"type": "http",
"url": "https://api.nitrosend.com/mcp"
}
}
}
OAuth sign-in happens automatically on first use.
Nitrosend is an official connector in the Claude directory. Go to Settings → Connectors, search for Nitrosend and click Connect, or open https://claude.ai/directory/nitrosend. Sign in when prompted; no URL to paste and no API key.
If your Claude organisation restricts directory connectors, ask an organisation owner to enable Nitrosend or to add it as a custom connector with:
https://api.nitrosend.com/mcp
Prefer a one-click desktop extension? mcpb/ packages the same remote server as a Claude Desktop extension (.mcpb) with browser OAuth. See mcpb/README.md to install or build it.
Nitrosend is an official plugin on the Cursor Marketplace. In Cursor's Agent chat, run:
/add-plugin nitrosend
Or click Add to Cursor on the Marketplace listing. The first Nitrosend tool call opens the OAuth sign-in; no API key, no config file.
Prefer to manage the server yourself? Add to ~/.cursor/mcp.json (all projects) or .cursor/mcp.json (this repo):
{
"mcpServers": {
"nitrosend": {
"url": "https://api.nitrosend.com/mcp"
}
}
}
Cursor handles Nitrosend as a direct remote MCP server. On first use, it prompts for OAuth sign-in.
Add to .vscode/mcp.json:
{
"servers": {
"nitrosend": {
"type": "http",
"url": "https://api.nitrosend.com/mcp"
}
}
}
VS Code handles OAuth automatically via its built-in MCP auth flow.
Add to ~/.codeium/windsurf/mcp_config.json:
{
"mcpServers": {
"nitrosend": {
"serverUrl": "https://api.nitrosend.com/mcp"
}
}
}
Windsurf supports OAuth natively for HTTP servers — sign in via your browser on first use.
codex mcp add nitrosend --url https://api.nitrosend.com/mcp
codex mcp login nitrosend
Or add to ~/.codex/config.toml:
[mcp_servers.nitrosend]
url = "https://api.nitrosend.com/mcp"
Then run codex mcp login nitrosend to authenticate via OAuth.
Add to your Zed settings:
{
"context_servers": {
"nitrosend": {
"url": "https://api.nitrosend.com/mcp"
}
}
}
If your client supports HTTP/SSE transport, point it at:
https://api.nitrosend.com/mcp
If your client only supports stdio, use this bridge package with an API key:
{
"command": "npx",
"args": ["-y", "@nitrosend/mcp"],
"env": {
"NITROSEND_API_KEY": "nskey_live_..."
}
}
Get your API key at Settings → API Keys in the Nitrosend dashboard.
Once connected, your agent can:
For OAuth connections, Nitrosend stores both the current account and the current brand per user and OAuth application. Ask your agent to call nitro_get_status first: it shows current_account and available_accounts alongside current_brand and available_brands.
If the task is for a different account than the one shown, call nitro_select_account with the target account_id. The switch takes effect on the next tool call. Then call nitro_select_brand to pick a brand within that account. The selected brand persists for this authenticated client across requests, reconnects, and token refreshes until you switch again.
nitro_select_account and nitro_select_brand only change the active MCP context. nitro_set_brand_kit edits the Brand Kit for the current brand, such as colors, logo, sender details, and voice.
API key connections are pinned to the API key's account and brand and cannot switch with nitro_select_account or nitro_select_brand. For stdio bridge deployments that must force a single brand, set NITROSEND_BRAND_SID; the bridge will send X-Brand-SID only when that variable is present.
| Variable | Description |
|---|---|
NITROSEND_API_KEY |
API key (nskey_live_...) — for stdio transport |
NITROSEND_BEARER_TOKEN |
OAuth bearer token — alternative to API key |
NITROSEND_BRAND_SID |
Optional fixed-brand override for the stdio bridge. Sends X-Brand-SID only when set. |
NITROSEND_API_URL |
Override API endpoint (default: https://api.nitrosend.com/mcp) |
MIT