AGENTS.md · mealie-recipes/mealie
AGENTS.md from mealie-recipes/mealie: 239 lines covering Project Overview, Architecture & Key Patterns, Essential Commands (via Task/Taskfile.yml), Critical…
Template by mealie-recipes
Chalk Labs
Need help deploying AGENTS.md · mealie-recipes/mealie?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode, Claude Code. Read the repository before you give any tool access to your data.
curl -fsSL https://raw.githubusercontent.com/mealie-recipes/mealie/mealie-next/AGENTS.md -o AGENTS.md
Read it and adapt it to your project before use.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: prompt-engineering, code-generation
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
What's in it
An AGENTS.md file: instructions coding agents (Codex, Cursor, Copilot, Gemini CLI and others) read before working in a repo.
- For
- Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode, Claude Code
- Length
- 239 lines
- File
AGENTS.md
Sections
- Project Overview
- Architecture & Key Patterns
- Essential Commands (via Task/Taskfile.yml)
- Critical Development Practices
- Pull Request Best Practices
- Common Gotchas
- Key Files to Reference
- Additional Resources
First lines (shown under the repo's AGPL-3.0 license)
# Mealie Development Guide for AI Agents
## Project Overview
Mealie is a self-hosted recipe manager, meal planner, and shopping list application with a FastAPI backend (Python 3.14) and Nuxt 4 frontend (Vue 3 + TypeScript). It uses SQLAlchemy ORM with support for SQLite and PostgreSQL databases.
**Development vs Production:**
- **Development:** Frontend (port 3000) and backend (port 9000) run as separate processes
- **Production:** Frontend is statically generated and served via FastAPI's SPA module (`mealie/routes/spa/`) in a single container
## Architecture & Key Patterns
### Backend Architecture (mealie/)
**Repository-Service-Controller Pattern:**
- **Controllers** (`mealie/routes/**/controller_*.py`): Inherit from `BaseUserController` or `BaseAdminController`, handle HTTP concerns, delegate to services
- **Services** (`mealie/services/`): Business logic layer, inherit from `BaseService`, coordinate repos and external dependencies
- **Repositories** (`mealie/repos/`): Data access layer using SQLAlchemy, accessed via `AllRepositories` factory
- Get repos via dependency injection: `repos: AllRepositories = Depends(get_repositories)`
- All repos scoped to group/household context automatically
**Route Organization:**
- Routes in `mealie/routes/` organized by domain (auth, recipe, groups, households, admin)
- Use `APIRouter` with FastAPI dependency injection
- Apply `@router.get/post/put/delete` decorators with Pydantic response models
- Route controllers use `HttpRepo` mixin for common CRUD operations (see `mealie/routes/_base/mixins.py`)
**Schemas & Type Generation:**
- Pydantic schemas in `mealie/schema/` with strict separation: `*In`, `*Out`, `*Create`, `*Update` suffixes
- Auto-exported from submodules via `__init__.py` files (generated by `task dev:generate`)
- TypeScript types auto-generated from Pydantic schemas - **never manually edit** `frontend/app/lib/api/types/`
**Database & Sessions:**
- Session management via `Depends(generate_session)` in FastAPI routes
- Use `session_context()` context manager in services/scripts
- SQLAlchemy models in `mealie/db/models/`, migrations in `mealie/alembic/`
- Create migrations: `task py:migrate -- "description"`
### Frontend Architecture (frontend/)
**Component Organization (strict naming conventions):**
- **Domain Components** (`components/Domain/`): Feature-specific, prefix with domain (e.g., `AdminDashboard`)
- **Global Components** (`components/global/`): Reusable primitives, prefix with `Base` (e.g., `BaseButton`)
- **Layout Components** (`components/Layout/`): Layout-only, prefix with `App` if props or `The` if singleton
- **Page Components** (`components/` with page prefix): Last resort for breaking up complex pages
**API Client Pattern:**
- API clients in `frontend/app/lib/api/` extend `BaseAPI`, `BaseCRUDAPI`, or `BaseCRUDAPIReadOnly`
- Types imported from auto-generated `frontend/app/lib/api/types/` (DO NOT EDIT MANUALLY)
- Composables in `frontend/app/composables/` for shared state and API logic (e.g., `use-mealie-auth.ts`)
- Use `useAuthBackend()` for authentication state, `useMealieAuth()` for user management
**State Management:**
- Nuxt 4 composables for state (no Vuex)
- Auth state via `use-mealie-auth.ts` composable
- Prefer composables over global state stores
## Essential Commands (via Task/Taskfile.yml)
**Development workflow:**
```bash
task setup # Install all dependencies (Python + Node)
task dev:services # Start Postgres & Mailpit containers
task py # Start FastAPI backend (port 9000)
task ui # Start Nuxt frontend (port 3000)
task docs # Start Zensical documentation server
```
**Code generation (REQUIRED after schema changes):**
```bash
task dev:generate # Generate TypeScript types, schema exports, test helpers
```
**Testing & Quality:**
```bash
task py:test # Run pytest (supports args: task py:test -- -k test_name)
task py:check # Format + lint + type-check + test (full validation)
task py:format # Ruff format
task py:lint # Ruff check
task py:mypy # Type checking
task ui:test # Vitest frontend tests
task ui:check # Frontend lint + test
```
**Database:**
```bash
task py:migrate -- "description" # Generate Alembic migration
task py:postgres # Run backend with PostgreSQL config
```
**Docker:**
```bash
task docker:prod # Build and run production Docker compose
```
## Critical Development Practices
### Python Backend
1. **Always use `uv` for Python commands** (not `python` or `pip`):
```bash
uv run python mealie/app.py
uv run pytest tests/
```
2. **Type hints are mandatory:** Use mypy-compatible annotations, handle Optional types explicitly
3. **Dependency injection pattern:**
```python
from fastapi import Depends
from mealie.repos.all_repositories import get_repositories, AllRepositories
def my_route(repos: AllRepositories = Depends(get_repositories), user: PrivateUser = Depends(get_current_user)):
recipe = repos.recipes.get_one(recipe_id)
```
4. **Settings & Configuration:**
- Get settings: `settings = get_app_settings()` (cached singleton)
- Get directories: `dirs = get_app_dirs()`
- Never instantiate `AppSettings()` directly
5. **Testing:**
- Fixtures in `tests/fixtures/`
- Use `api_client` fixture for integration tests
- Follow existing patterns in `tests/integration_tests/` and `tests/unit_tests/`
### Frontend
1. **Run code generation after backend schema changes:** `task dev:generate`
2. **TypeScript strict mode:** All code must pass type checking
3. **Component naming:** Follow strict conventions (see Architecture section above)
4. **API calls pattern:**
```typescript
const api = useUserApi();
const recipe = await api.recipes.getOne(recipeId);
```
5. **Composables for shared logic:** Prefer composables in `composables/` over inline code duplication
6. **Translations:** Only modify `en-US` locale files when adding new translation strings - other locales are managed via Crowdin and **must never be modified** (PRs modifying non-English locales will be rejected)
### Cross-Cutting Concerns
1. **Code generation is source of truth:** After Pydantic schema changes, run `task dev:generate` to update:
- TypeScript types (`frontend/app/lib/api/types/`)
- Schema exports (`mealie/schema/*/__init__.py`)
- Test data paths and routes
2. **Multi-tenancy:** All data scoped to **groups** and **households**:
- Groups contain multiple households
- Households contain recipes, meal plans, shopping lists
- Repositories automatically filter by group/household context
3. **Pre-commit hooks:** Install via `task setup:py`, enforces Ruff formatting/linting
4. **Testing before PRs:** Run `task py:check` and `task ui:check` before submitting PRsSignals
- Stars
- 13,670
- Forks
- 1,601
- Open issues
- 68
- Last commit (this folder)
- Sep 23, 2026 (1 month ago)
- Last release
- Oct 17, 2026 (13 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- AGPL-3.0
- Repository
- github.com/mealie-recipes/mealie
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get AGENTS.md · mealie-recipes/mealie running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
AGENTS.md · mealie-recipes/mealie. Agent Mart. https://agentmart.dev/templates/mealie-recipes-mealie-agents-md (last updated 2026-10-24)