Back to all glossary terms

MCP

An open protocol that lets AI agents discover and call a server's tools at runtime over JSON-RPC, instead of relying on integration code written in advance.

FBFenwei BianAuthorVZVictor ZhouEditorAugust 3, 2026
  • glossary

MCP, or the Model Context Protocol, is an open protocol — released by Anthropic on 25 November 2024 — that lets an AI agent discover and call a server's tools at runtime rather than through integration code written in advance. It runs over JSON-RPC 2.0: a tools/list request returns the tools the caller may invoke, each carrying a name, a description, and an inputSchema that must "be a valid JSON Schema object" (2020-12 by default), plus an optional outputSchema for the result; tools/call then invokes one. Servers are reached over stdio, where "the client launches the MCP server as a subprocess", or over Streamable HTTP, a single POST endpoint; HTTP servers that support authorization act as OAuth 2.1 resource servers, and clients "MUST implement PKCE". Failures split two ways — JSON-RPC errors for malformed requests, and tool execution errors returned with isError: true, giving the model "actionable feedback" to self-correct against. As of the 2026-07-28 revision there is no initialize handshake — "There is no negotiation handshake"; each request declares its own protocol version, and "MCP has no protocol-level session". The protocol itself "has no concept of a state handle"; as non-normative guidance, a server needing state across calls should return an explicit handle and accept it as an argument on later calls. Compare EPP, the protocol registrars use with registries, and x402 for agent payments; Why MCP Exists When We Already Have REST APIs and CLIs covers the design rationale.

Related keywords

  • Model Context Protocol
  • MCP server
  • tools/list
  • JSON-RPC 2.0
  • agent tool calling
  • MCP
  • what is MCP
  • MCP meaning
  • MCP definition

Contributors

Fenwei Bian

Software Developer & Writer • Namefi
Author

Fenwei Bian is a software developer in her thirties who spends her working hours in pull requests and her weekends with her hands in soil or sawdust. Years of open source on GitHub taught her that names are interfaces: a good one is clear, honest about what it does, and kind to whoever has to use it next.

She gardens because it rewards patience and punishes wishful thinking, and she does woodwork because a joint either fits or it doesn't. Both habits show up in how she writes about naming — measure twice, check the source, and don't sand over a rough spot and hope no one notices.

For Namefi she writes about how domain markets actually move, the practical trade-offs of tokenizing and flipping names, and picking a domain you'll still be glad you own in twenty years.

GitHub

Victor Zhou

Founder & Standards Editor • Namefi
Editor

Victor Zhou is a technology founder and standards editor focused on digital identity and trust. He founded Namefi, edits Ethereum Improvement Proposals, and previously led smart-contract architecture work at Google Labs.

His work sits at the intersection of naming, ownership, and the systems people use to establish identity online. That perspective makes him especially interested in the way names move between personal meaning, public recognition, and digital infrastructure.

For Namefi, Victor edits and writes about domains as durable digital identity: how names become ownable onchain assets, how tokenization changes custody and trust, and what naming can learn from the systems people use to establish identity online.

WebsiteBlogGitHubLinkedIn

Related guides