Model Context Protocol
MCP is an open standard that lets AI assistants talk to your tools and data through one common interface. An MCP server is a small adapter that wraps a system — Figma, Sentry, a database — and exposes it in a way any MCP-capable assistant can use.
Think of MCP as a USB-C port for AI apps: one standard plug. Build a server once, and every assistant that speaks MCP can connect — no custom integration per app.
How a request flows
The assistant never touches your system directly — it speaks MCP to the server, which does the real work. Press play to watch a single request travel end to end.
1/7You ask the assistant a question it can't answer on its own.
What a server exposes
Every MCP server offers some mix of three primitives.
Tools
Actions the model can invoke — search issues, create a record, run a query.
tools/call → create_taskResources
Read-only data the model can pull in for context — files, records, documents.
resources/read → file://…Prompts
Reusable, parameterised prompt templates the server offers to the user or model.
prompts/get → summarizeHow it connects
An MCP server runs in one of two places.
Local (stdio)
The client launches the server as a process on your machine and talks to it over stdin/stdout. Great for local tools and files.
Remote (HTTP / SSE)
The server runs as a hosted URL the client connects to over HTTP. Auth is usually OAuth. Every server in this directory is remote.
Ready to connect one?
Browse the MCP servers Signifly uses, with setup for your client.