Omentir — Claude Code MCP server: LinkedIn lead search and outreach
Hooks your agent into an Omentir workspace so it can surface LinkedIn prospects, draft messages and run campaigns at a pace that looks human.
claude mcp add --transport http omentir https://omentir.com/api/agent/v1/mcpWhat Omentir does
The endpoint lives at https://omentir.com/api/agent/v1/mcp and speaks HTTP. Behind it sits lead search on LinkedIn, contextual message drafts, campaigns with slow sending, and reply handling. Ask the assistant for a standard lead finder, or for a steal_customers run that takes competitor company URLs as input. With several companies on the account, have it list workspaces and switch.
Chat apps connect through Connectors with a sign-in and an approval step, no API key involved. Cursor, Claude and scripts need a key from the API page, sent as Authorization: Bearer <token> — the same token covers the REST routes under /api/agent/v1/*. Your LinkedIn password never goes to the agent; the account is linked inside Omentir itself.
The server belongs to an app you can host yourself, and that route is not offline: bring your own Firebase, Unipile and access to Google AI, either a Gemini key or a Vertex AI project in Google Cloud. Self-hosting drops the Omentir subscription, yet those providers can still bill you. Local setup wants Git plus Docker Compose or Bun 1.3, and .env has to carry the required secrets. Miss LOCAL_SESSION_SECRET at the right length and the app refuses to boot.
What it helps with
- find LinkedIn people who match an ideal-customer profile
- draft a first message tuned to someone's profile context
- launch a campaign with restrained sending pace
- sort incoming replies and hand them to the agent
- turn competitor URLs into a prospect list via
steal_customers
Example prompts for the agent
- Connect Omentir and list my workspaces
- Find 20 fintech contacts on LinkedIn and write drafts for each
- Create a
steal_customersagent for these competitor URLs
How to install Omentir
Platform CLI: remote server
claude mcp add --transport http omentir https://omentir.com/api/agent/v1/mcpPlatform CLI: remote server
codex mcp add omentir --url https://omentir.com/api/agent/v1/mcpConfig file
{
"mcpServers": {
"omentir": {
"url": "https://omentir.com/api/agent/v1/mcp"
}
}
}Config file
{
"servers": {
"omentir": {
"type": "http",
"url": "https://omentir.com/api/agent/v1/mcp"
}
}
}Platform CLI: remote server
gemini mcp add -t http omentir https://omentir.com/api/agent/v1/mcpStep by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about Omentir
How do I install Omentir in Claude Code?
Run claude mcp add --transport http omentir https://omentir.com/api/agent/v1/mcp. Then generate a key on the API page and pass it in the Authorization: Bearer <token> header.
What does Omentir do?
It finds LinkedIn prospects, writes context-aware outreach and runs campaigns while sorting replies. You drive it from your AI agent rather than handing over your LinkedIn password.
Can Omentir run on my own server?
Yes, the app is self-hostable, though it leans on outside services. You supply Firebase, Unipile, and either a Gemini key or a Google Cloud Vertex AI project; those can cost money.
