Model Context Protocol
An open protocol connecting AI systems to external data sources and tools through a client-server interface, in which a server exposes a declared set of resources, tools and prompts to a host application's client.
Why it matters
MCP describes capability, not permission. It answers what an agent can reach, and is silent on what it may commit, under whose authority, and up to what limit.
What it is not
These are routinely confused with Model Context Protocol. The distinctions are not pedantic — each one has consequences for how a system is governed.
MCP connects an agent to tools and data. A2A connects agents to each other. They operate at different layers and are complementary, not competing.
Relationships
Typed edges into the rest of the ontology. These are what make the canon traversable rather than merely readable.
| Verb | Target | Meaning |
|---|---|---|
produces | Tool Manifest | The subject emits the object as an output. |
notEquivalentTo | Agent2Agent | The two are routinely conflated and are distinct. |
dependsOn | MCP | The subject cannot function correctly without the object. |
Record
| Canonical identifier | QIS-TERM-00017 |
| Status | Canonical industry term |
| Adoption | Widely used |
| Domain · Layer | Protocol · Intelligence |
| Origin | Anthropic; published as an open specification. |
| Semantic aliases | None recorded. |
| First published | 2026-08-02 |
| Last reviewed | 2026-08-02 · 180-day cycle |
Cite the identifier, not the URL. Identifiers are stable; URLs may change. This entry is free to read, quote and index under the dual license. Corrections to [email protected] are published.