# Concepts

These pages describe the core concepts of the platform and how they fit together. Sources are deployed to generate tool definitions, tool definitions are curated into toolsets, toolsets are served as MCP servers, and tool variations refine how tools appear to LLMs along the way.

## [Tool sources](/docs/ai-control-plane/reference/concepts/tool-sources)

The inputs that describe available functionality: OpenAPI documents, Gram Functions, and external MCP servers. Uploading a source is the starting point for creating tools.

## [Deployments](/docs/ai-control-plane/reference/concepts/deployments)

Immutable snapshots of a project's sources and the tools generated from them. Every source change produces a new deployment, and the most recent successful deployment serves the project's tools.

## [Tool definitions](/docs/ai-control-plane/reference/concepts/tool-definitions)

The output of processing a source: one definition per API operation, function tool, or external MCP tool, containing the metadata an LLM needs and the configuration the platform uses to execute the tool.

## [Toolsets](/docs/ai-control-plane/reference/concepts/toolsets)

Curated collections of tool definitions scoped to a specific goal or team. Each toolset can be served as a hosted MCP server.

## [Tool variations](/docs/ai-control-plane/reference/concepts/tool-variations)

Overrides for a tool's name, description, tags, annotations, and confirmation prompts — a prompt-engineering layer that improves how LLMs select and use tools without changing the underlying source.