Glossary category

Tools and protocols

The interfaces and contracts that let AI systems act on software and external services.

What belongs here

Tool calling, capabilities, gateways, protocols, schemas, and interoperability boundaries.

What belongs elsewhere

The reasoning policy that chooses a tool and the security policy that authorizes it.

Adjacent categories

Definitions in this category

stable

Agent Card

An A2A discovery document that describes an agent's identity, supported interfaces, capabilities, skills, and security schemes.

Agent2Agent Protocol (A2A)

An open protocol for discovering and collaborating with independent agent services through messages, tasks, artifacts, and updates.

Application programming interface (API)

A documented interface through which one software component requests data or behavior from another.

Capability

A bounded action or class of action that a system component can perform, usually paired with explicit authority and scope.

Ephemeral environment

A short-lived execution environment created for a task or run and destroyed when that work ends.

Execution lineage

A reconstructable chain linking an outcome to the intent, identities, system versions, inputs, actions, evidence, and state changes that produced it.

Latency

Elapsed time between a defined starting event and a defined response or completion point.

MCP capability negotiation

The MCP initialization exchange in which a client and server declare the optional protocol features they support.

MCP client

The MCP protocol component created by a host to establish and manage a connection with one server.

MCP gateway

An intermediary that fronts one or more MCP servers and centralizes routing, authentication, policy, limits, or audit.

MCP host

The AI application that coordinates users, models, policy, and one or more MCP server connections.

MCP prompt

A reusable message template exposed by an MCP server for a user to select and supply to a model interaction.

MCP resource

Data exposed by an MCP server under a URI so a host can retrieve it and decide whether to place it in model context.

MCP server

A program or service that exposes tools, resources, prompts, or other negotiated MCP capabilities to a connected client.

MCP tool

An executable operation exposed by an MCP server for model discovery and invocation through a defined schema.

Model Context Protocol (MCP)

An open client-server protocol for connecting AI applications to external tools and context through standard discovery and interaction primitives.

Rate limit

A constraint on the requests, tokens, compute, or actions allowed within a defined time window.

Sandbox

An execution environment that enforces boundaries around files, processes, networks, credentials, and other resources.

Throughput

The amount of defined work a system completes during a unit of time.

Tool

An executable capability an AI application or agent can invoke to retrieve information, run computation, or change state.