AI Integration

Ecotone provides multiple ways for AI assistants and code editors to access up-to-date documentation. This ensures AI tools give accurate, current answers about Ecotone Framework.

Ecotone Skills

Ecotone provides Agent Skills that give AI coding agents deep knowledge of Ecotone's patterns, conventions, and best practices. Skills follow the open Agent Skills standardarrow-up-right (SKILL.md), making them compatible with a growing number of coding agents.

When installed, Ecotone Skills are automatically activated based on context — the AI agent detects when you're working with Ecotone and loads relevant guidance for Message Handlers, Aggregates, Event Sourcing, Asynchronous Processing, Workflows, Testing, and more.

Skills repository: github.com/ecotoneframework/skillsarrow-up-right

Claude Code

Install Ecotone Skills as a plugin:

/plugin install ecotone-skills@ecotone

This installs skills, slash commands, and subagents tailored for Ecotone development. Manage the plugin with:

/plugin disable ecotone-skills@ecotone
/plugin enable ecotone-skills@ecotone
/plugin uninstall ecotone-skills@ecotone

Cursor

Cursor supports the SKILL.md standard natively. Install using the Vercel skills CLI:

npx skills add ecotoneframework/skills -a cursor

Or place Ecotone skills manually in your project:

Windsurf

Windsurf's Cascade agent supports Agent Skills. Place skills in:

GitHub Copilot

GitHub Copilot supports Agent Skills in VS Code. Place skills in:

Cline

Cline supports skills since version 3.48.0. Enable them in Settings → Features → Enable Skills. Skills are placed in the standard .claude/skills/ directory.

Augment Code

Augment Code discovers skills automatically from multiple directories:

Kiro

Kiro supports skills at user and workspace level:

OpenAI Codex CLI

Codex CLI reads skills from ~/.codex/skills/ and must be started with the --enable skills flag.

MCP Server (Model Context Protocol)

The MCP server provides direct access to Ecotone documentation for AI assistants that support the Model Context Protocol.

MCP Server URL:

VSCode / Cursor Integration

Click the link below to instantly add Ecotone MCP to your VSCode or Cursor:

Install Ecotone MCP in VSCodearrow-up-right

Or manually add to your MCP settings:

Claude Desktop

Add to your Claude Desktop MCP configuration (claude_desktop_config.json):

Context7 Integration

Context7arrow-up-right provides up-to-date documentation for LLMs and AI code editors. Ecotone documentation is available through Context7.

Using Context7 MCP

Add Context7 MCP to access Ecotone and other library documentation:

Then query Ecotone documentation through your AI assistant with Context7 enabled.

LLMs.txt

Ecotone follows the llmstxt.orgarrow-up-right specification for AI-friendly documentation.

Available files:

File
Description

Concise overview with key links

Extended documentation with code examples

These files are automatically discovered by AI crawlers via robots.txt.

Best Practices for AI-Assisted Development

When using AI assistants with Ecotone:

  1. Enable MCP - Connect the Ecotone MCP server for real-time documentation access

  2. Reference specific features - Ask about specific patterns like "Ecotone Saga" or "Ecotone Event Sourcing"

  3. Include context - Mention your framework (Symfony/Laravel) for framework-specific guidance

  4. Verify code - Always test AI-generated code against the official documentation

Supported AI Tools

Ecotone documentation integrates with:

Tool
MCP
Skills
llms.txt

Claude Code

Yes

Yes (/plugin install)

Yes

Cursor

Yes

Yes (.cursor/skills/)

Yes

Windsurf

Yes

Yes (.windsurf/skills/)

Yes

GitHub Copilot

Yes

Yes (.github/skills/)

Yes

Cline

Yes

Yes (Settings → Enable Skills)

Augment Code

Yes (.augment/skills/)

Kiro

Yes (.kiro/skills/)

OpenAI Codex CLI

Yes (~/.codex/skills/)

ChatGPT

Yes

Context7-enabled tools

Yes

Resources

Materials

Last updated

Was this helpful?