Skip to main content

Overview

Enferno ships three pieces of context for coding agents:
  • AGENTS.md: a short, always-loaded guide with the stack, layout, commands, routes and hard rules. Claude Code, Codex, Cursor, GitHub Copilot and most other agents read it automatically.
  • The enferno-dev skill in .agents/skills/enferno-dev/: full code patterns for models, APIs, Vue pages, dialogs and migrations. Agents load it only when a task needs it, so it costs nothing the rest of the time.
  • flask mcp: a local, read-only MCP server with the live routes, schema and row counts, migration revision and audit log. Install it with uv sync --extra dev --extra mcp; .mcp.json registers it, and your agent asks before connecting.
.agents/skills/ is the shared location for Codex, Cursor, Copilot, Gemini CLI, OpenCode and others. .claude/skills is a symlink to it, so Claude Code sees the same skill.

Using It

Open the project in your agent and ask for features in plain language:

Using the Skill in Another Project

Install the skill into any project with the skills CLI:

Customizing for Your Project

Add project-specific rules to AGENTS.md and keep it short. Put longer patterns in a skill under .agents/skills/<name>/SKILL.md so they load only when relevant.