Skip to content

Glossary

Model Context Protocol

Model Context Protocol (MCP) is an open standard for connecting AI clients to tools and data. Clients call a server over a shared protocol instead of inventing a one-off integration per product.

Before MCP, every AI product needed a custom plug-in path. MCP standardizes how a client discovers tools, calls them, and receives structured results.

Zevari hosts a LinkedIn MCP server so your agent can drive the same six specialists the Zevari app runs: prospecting, signals, ICP scoring, drafting, follow-up, and reply qualification. Zevari hosts the LinkedIn connection, state, and safety controls, and every send waits for your approval until a specialist has earned autopilot.

Questions

FAQ

Is MCP only for Claude?

No. Any MCP-capable client can connect to a hosted MCP server. Zevari's LinkedIn MCP works with Claude Code, Codex, ChatGPT, OpenClaw, Grok Bot in Cursor, and other MCP clients - or the same engine over REST.

Does adding an MCP endpoint send LinkedIn messages by itself?

No. Connecting the endpoint only makes tools available. In Zevari, every write still waits for your approval by default.

Keep going

MCP endpoint free to add. Operator $87/mo, extra accounts $37, $497/yr First 100 Club, 30-day money-back. Approval gate on every write.