SourceDocker Blog•Stalecollected in 20m
Custom MCP Catalogs and Profiles for Enterprise AI
Learn how to standardize and scale AI tool distribution across your enterprise using Docker's new MCP management tools.
30-Second TL;DR
What Changed
Custom MCP Catalogs enable centralized curation and distribution of approved AI tools.
Why It Matters
This update significantly reduces the friction of deploying AI agents in enterprise environments by providing a structured way to manage tool dependencies and configurations.
What To Do Next
Review your current MCP server deployment workflow and create a standardized Profile to streamline your team's local development environment.
Who should care:Enterprise & Security Teams
Key Points
- •Custom MCP Catalogs enable centralized curation and distribution of approved AI tools.
- •MCP Profiles simplify the configuration and execution of AI servers for individual developers.
- •Enhances enterprise-wide adoption of MCP by streamlining tool management and governance.
Deep Insight
Background and context from public sources — not the original article. 17 sources cited.
Enhanced Key Takeaways
- •The Model Context Protocol (MCP) is an open, JSON-RPC-based standard, initially published by Anthropic in November 2024, designed to enable seamless integration between LLM applications and external data sources and tools, thereby addressing the 'N x M integration problem' by standardizing tool discovery and invocation.
- •Custom Catalogs allow organizations to curate specific subsets of approved MCP servers, including proprietary internal servers, and host them in their own container registries, which enhances security, compliance, and control over the AI tools available to development teams beyond Docker's public catalog.
- •MCP Profiles centralize server configurations into named collections for different projects, enabling developers to easily switch between tailored toolsets and share these configurations with their teams, significantly reducing individual setup overhead and ensuring consistent development environments.
- •Docker's MCP tooling, including the MCP Gateway, leverages Open Container Initiative (OCI) artifacts for catalogs and profiles, which facilitates immutable versioning, GitOps integration, and consistent security scanning and provenance tracking for enterprise-grade governance.
- •The MCP Gateway acts as a single connection point for AI clients to access multiple MCP servers, handling authentication, lifecycle management, and supporting dynamic discovery and on-demand instantiation of servers from curated catalogs.
Technical Deep Dive
- The Model Context Protocol (MCP) is an open and evolving JSON-RPC-based standard.
- It enables AI applications to discover and interact with external tools, reusable prompts, and data resources from remote MCP servers through a stateful session.
- MCP defines three core primitives that servers can expose: Tools (executable functions), Resources (read-only contextual data like files or database schemas), and Prompts (reusable templates for structured interactions).
- The architecture is client-server, where the Large Language Model (LLM) or AI agent acts as the client, and enterprise tools or data systems function as MCP servers.
- Communication between clients and servers typically occurs over WebSocket or HTTP.
- MCP draws inspiration from the Language Server Protocol (LSP) for standardizing how development tools integrate with programming languages.
- Docker packages MCP servers as OCI artifacts (container images) for distribution via Docker Hub and custom registries, ensuring versioning, provenance, and security updates.
- The MCP Gateway serves as a centralized connection point, routing client requests to the appropriate MCP server and managing authentication and server lifecycle.
- Dynamic MCP is a feature that allows AI agents to discover and add MCP servers on-demand during conversations, searching within the defined catalogs.
- MCP Apps (formerly mcp-ui), formalized in early 2026, is an official extension that standardizes the delivery of interactive user interfaces (e.g., React-based dashboards, forms) from MCP servers to host AI applications.
Future ImplicationsAI analysis grounded in cited sources
Enterprise AI development will experience accelerated adoption of standardized tooling.
Docker's solution simplifies the secure and governed distribution of AI tools, reducing integration friction and increasing efficiency for large organizations.
The 'N x M integration problem' for AI tools will be significantly mitigated across enterprises.
MCP, combined with Docker's management features, provides a universal interface for AI applications to interact with diverse external systems, drastically reducing the need for custom integration efforts.
AI agent workflows will become more robust, auditable, and compliant in enterprise environments.
Leveraging OCI artifacts for MCP catalogs and profiles enables immutable versioning, security scanning, and clear provenance, which are crucial for governance and regulatory compliance.
Timeline
2023-Q4
Docker announced the GenAI Stack with partners Neo4j, LangChain, and Ollama, expanding its AI capabilities.
2024-11
Anthropic introduced the Model Context Protocol (MCP) as an open standard to standardize AI system integration.
2025-01
Docker launched Gordon, an AI agent integrated into Docker Desktop to assist with Docker-specific tasks.
2025-04
Docker Desktop 4.40 introduced Model Runner (Beta) for local LLM execution and enhanced Docker AI Agent with MCP integration.
2025-11
Docker Desktop 4.50 included early support for Model Context Protocol (MCP) tooling and Experimental Dynamic MCP.
2026-05
Docker announced the general availability of Custom Catalogs and Profiles for MCP servers.
- 2023-Q4Docker announced the GenAI Stack with partners Neo4j, LangChain, and Ollama, expanding its AI capabilities.
- 2024-11Anthropic introduced the Model Context Protocol (MCP) as an open standard to standardize AI system integration.
- 2025-01Docker launched Gordon, an AI agent integrated into Docker Desktop to assist with Docker-specific tasks.
- 2025-04Docker Desktop 4.40 introduced Model Runner (Beta) for local LLM execution and enhanced Docker AI Agent with MCP integration.
- 2025-11Docker Desktop 4.50 included early support for Model Context Protocol (MCP) tooling and Experimental Dynamic MCP.
- 2026-05Docker announced the general availability of Custom Catalogs and Profiles for MCP servers.
Sources (17)
Factual claims are grounded in the sources below. Forward-looking analysis is AI-generated interpretation.
1Google Search Sourcevertexaisearch.cloud.google.com2Google Search Sourcevertexaisearch.cloud.google.com3Google Search Sourcevertexaisearch.cloud.google.com4Google Search Sourcevertexaisearch.cloud.google.com5Google Search Sourcevertexaisearch.cloud.google.com6Google Search Sourcevertexaisearch.cloud.google.com7Google Search Sourcevertexaisearch.cloud.google.com8Google Search Sourcevertexaisearch.cloud.google.com9Google Search Sourcevertexaisearch.cloud.google.com10Google Search Sourcevertexaisearch.cloud.google.com11Google Search Sourcevertexaisearch.cloud.google.com12Google Search Sourcevertexaisearch.cloud.google.com13Google Search Sourcevertexaisearch.cloud.google.com14Google Search Sourcevertexaisearch.cloud.google.com15Google Search Sourcevertexaisearch.cloud.google.com16Google Search Sourcevertexaisearch.cloud.google.com17Google Search Sourcevertexaisearch.cloud.google.com
Weekly AI Recap
Read this week's curated digest of top AI events →
AI-curated news aggregator. All content rights belong to original publishers.
Original source: Docker Blog ↗
This is a summary, not the original. Read the source, or get the weekly briefing.
The weekly digest
One email a week. Unsubscribe anytime.