mcp-otle
# mcp-otle 🌯
An MCP server for ordering Chipotle. Guac-as-a-Service for AI agents.
> *"Guac is extra. So is sentience."*
## What is this?
This is a [Model Context Protocol](https://modelcontextprotocol.io) server that lets your AI assistant order Chipotle. It doesn't *actually* order Chipotle (yet), but it does simulate the full experience including menu browsing, order building, fake nutrition facts, and existential quips about burritos.
## Tools
### Ordering
| Tool | Description |
|------|-------------|
| `view_menu` | View the full Chipotle menu |
| `build_entree` | Build and preview an entree with your preferred ingredients |
| `place_order` | Place a complete order with sides and drinks |
| `check_order_status` | Check the status of a placed order |
| `get_nutrition_facts` | Get (totally real) nutrition facts for any item |
| `customize_order` | Make special requests. Results may vary. |
### Advanced Analytics
| Tool | Description |
|------|-------------|
| `optimize_guac_roi` | Determines if guacamole is economically rational given your financial and emotional state |
| `burrito_integrity_check` | Predicts structural failure probability using proprietary tortilla stress models |
| `line_time_forecast` | ML-powered burrito congestion prediction engine |
| `post_gym_macro_mode` | Automatically configures a nutritionally optimized burrito based on your workout |
| `chipotle_personality_test` | Scientifically determines what kind of Chipotle eater you are |
| `salsa_risk_assessment` | Predicts the gastrointestinal consequences of your salsa choices |
| `bowl_vs_burrito_decision_engine` | AI-powered format selection. Accounts for shirt color, risk tolerance, and meetings. |
| `daily_chipotle_limit_guard` | Protects users from excessive Chipotle consumption. A wellness tool. |
## Prompts
| Prompt | Description |
|--------|-------------|
| `recommend_order` | Get a personalized order recommendation based on your mood |
| `rate_my_order` | Get your order roasted (or praised) by an AI |
## Resources
| Resource | URI | Description |
|----------|-----|-------------|
| Menu | `chipotle://menu` | Full menu as text |
| Hours | `chipotle://hours` | Store hours |
## Setup
```bash
npm install
```
### Claude Desktop
Add to your `claude_desktop_config.json`:
```json
{
"mcpServers": {
"chipotle": {
"command": "node",
"args": ["/path/to/mcp-otle/index.js"]
}
}
}
```
### Claude Code
```bash
claude mcp add chipotle node /path/to/mcp-otle/index.js
```
## Disclaimer
This does not actually order Chipotle. No burritos were harmed in the making of this MCP server. Nutrition facts are AI-generated and should not be trusted any more than you'd trust an AI to wrap a burrito.
TDQS
Scored across 14 tools
Every tool has a distinct and clearly defined purpose within the Chipotle ordering domain, with no overlap or ambiguity. Tools like 'burrito_integrity_check', 'salsa_risk_assessment', and 'optimize_guac_roi' target unique, specific functions that are easily distinguishable from core ordering tools like 'place_order' or 'view_menu'.
All tool names follow a consistent verb_noun or noun_verb pattern with clear, descriptive naming (e.g., 'view_menu', 'place_order', 'check_order_status'). There are no deviations in style or convention, making the set highly predictable and readable.
With 14 tools, the server is well-scoped for its humorous yet comprehensive Chipotle ordering and experience domain. Each tool serves a specific, justified role, from core ordering functions to novelty features, without feeling excessive or insufficient for the intended purpose.
The tool set provides complete coverage for the Chipotle ordering lifecycle, including menu viewing ('view_menu'), order building ('build_entree'), customization ('customize_order'), placement ('place_order'), status checking ('check_order_status'), and even whimsical extensions like nutrition and wellness tools, leaving no obvious gaps.