AI tools and projects in the Drupal ecosystem

The Drupal community has been building quickly in the AI space, and the landscape is expanding fast. This page is a brief orientation to the categories of tooling available and where to find the most current information — because any comprehensive directory published here would be out of date within weeks.

If you want a community-maintained, comprehensive list that anyone can contribute to, awesome-drupal-ai is the place to start. It covers contributed modules, skill repositories, DDEV plugins, MCP servers, and more, and accepts contributions via pull request.

Categories of tooling

Agent skills and context files

Agent skills and context files are plain text Markdown files you place in your project root to tell AI coding tools about your specific codebase and conventions. Think of them as a persistent briefing document: instead of explaining "this is a Drupal project, follow these standards, prefer contrib over custom code" at the start of every session, you write it once and the tool reads it automatically. Skills are modular, shareable versions of these files built to a common standard so they work across different tools.

The Drupal community has produced a growing number of skills covering coding standards, module structure, security, contribution workflows, DDEV setup, and more. To browse what is available across all tools, skills.sh/?q=drupal lists installable Drupal skills following the agentskills.io open standard. On drupal.org specifically, ai_skills is the community's validated skill library, and Surge aggregates skills from multiple sources and generates context files via Composer. The ai_best_practices project is actively consolidating the best of these into a single Composer-installable package.

Drupal as an AI-capable site (the AI module ecosystem)

The AI module provides a unified abstraction layer for 48+ AI providers — OpenAI, Anthropic, Google, AWS Bedrock, Ollama, and many more — with submodules covering content generation, translation, semantic search, CKEditor integration, RAG, and agentic workflows. If you want to add AI capabilities to a Drupal site rather than use AI tools to build one, this is where to start. drupal.org/ai is the official entry point to this ecosystem.

MCP servers

Model Context Protocol (MCP) allows AI coding agents to interact with external tools and data sources in real time, rather than relying solely on static context files. For Drupal this works in two directions: connecting your coding agent to your running Drupal site (so it can read content types, run Drush commands, inspect configuration), and exposing your Drupal site as an MCP server for external AI systems to query.

drupalmcp.io is the dedicated hub for Drupal MCP documentation and tooling. Key modules include mcp_server (turns your Drupal site into an MCP endpoint) and mcp_tools (exposes Drupal to your coding agent via Drush). For a broader view of MCP servers with Drupal entries, glama.ai/mcp/servers and mcpservers.org are the main public registries.

Drupal.org contribution tooling

Drupal's community contribution workflow — issue queues, patches, merge requests — has its own emerging category of AI tooling. drupalorg-cli includes a --format=llm flag that formats output for agent consumption, making it easier for AI tools to work with Drupal.org issues and merge requests directly.

Specifications and standards

Understanding the underlying standards helps you evaluate tools and avoid vendor lock-in. The key ones to know:

• agentskills.io — the open SKILL.md format supported by tools including Claude Code, Cursor, GitHub Copilot, Gemini CLI, and Kiro
• AGENTS.md — a universal context file format supported by most major coding agents
• llmstxt.org — a standard for making your site's content discoverable to AI systems

A note on pace

This is one of the fastest-moving areas in Drupal development. Projects that are experimental today may be stable and widely adopted within months; others may be superseded. Where possible, the links above point to living community resources rather than static lists. If you find something missing or out of date, the awesome-drupal-ai repository accepts contributions, as does the ai_best_practices issue queue.