LLM Tool Demo

v 18.0 Third Party 8

Availability	Odoo Online Odoo.sh On Premise
Odoo Apps Dependencies	• CRM (crm) • Discuss (mail) • Calendar (calendar) • Contacts (contacts) • Invoicing (account)
Community Apps Dependencies	Show • LLM Tool • LLM Integration Base
Lines of code	2776
Technical Name	`llm_tool_demo`
License	LGPL-3
Website	https://github.com/apexive/odoo-llm

You bought this module and need support? Click here!

Availability	Odoo Online Odoo.sh On Premise
Odoo Apps Dependencies	• CRM (crm) • Discuss (mail) • Calendar (calendar) • Contacts (contacts) • Invoicing (account)
Community Apps Dependencies	Show • LLM Tool • LLM Integration Base
Lines of code	2776
Technical Name	`llm_tool_demo`
License	LGPL-3
Website	https://github.com/apexive/odoo-llm

Description
Documentation

Developer Reference

LLM Tool Demo

Demonstration of @llm_tool decorator usage.

Learn how to create custom AI tools with practical examples

6 Examples

Best Practices

Open Source

What is LLM Tool Demo?

Your guide to creating custom AI tools for Odoo

This module demonstrates how to create LLM tools using the @llm_tool decorator. It provides 6 example tools showing different decorator patterns and best practices for tool development. Use this as a reference when creating your own AI tools for Odoo.

Included Examples

6 example tools demonstrating different patterns

get_system_info

Simple read-only tool example showing basic decorator usage.

calculate_business_days

Utility tool with type hints for automatic schema generation.

create_lead_from_description

Business logic tool showing CRM integration patterns.

generate_sales_report

Complex reporting tool demonstrating Sales module integration.

legacy_example

Manual schema example for integrating with legacy code.

send_notification_to_user

User interaction tool showing Mail integration for notifications.

What You'll Learn

Best practices demonstrated in these examples

Tool Types

Read-only, destructive, and idempotent tool patterns for safe AI operations.

Type Hints

Use Python type hints for automatic schema generation and validation.

Business Logic

Integration with CRM, Sales, and Mail modules for real-world use cases.

Technical Details

Requirements and dependencies

Module Information

Dependencies:

llm_tool, crm, sale, mail

Category:

Productivity/LLM

Purpose:

Developer reference

License:

LGPL-3

How to Use

Install this module, explore the code in the models directory, and use the patterns in your own modules.

Related Modules

Explore other modules in the Odoo LLM suite

LLM Tool

The core tool framework that provides the @llm_tool decorator.

LLM Base

Core framework for LLM integration with provider management and model configuration.

LLM Thread

Conversational AI threads where tools can be executed.

LLM Assistant

Create assistants with specific tools enabled for their workflows.

OpenAI Provider

Connect to OpenAI GPT models with native function calling support.

MCP Server

Expose tools via Model Context Protocol for external AI clients.

LLM Tool Demo Module

Learn how to create LLM tools with examples.

Module Type: 📖 Demo/Tutorial (Tool Development)

Architecture

┌───────────────────────────────────────────────────────────────┐
│                     Consumers of Tools                        │
│  ┌─────────────┐  ┌─────────────┐  ┌─────────────────────┐   │
│  │llm_assistant│  │  llm_letta  │  │   llm_mcp_server    │   │
│  └──────┬──────┘  └──────┬──────┘  └──────────┬──────────┘   │
└─────────┼────────────────┼────────────────────┼──────────────┘
          └────────────────┼────────────────────┘
                           ▼
              ┌───────────────────────────────────────────┐
              │              llm_tool                     │
              │         (Tool Framework)                  │
              └─────────────────────┬─────────────────────┘
                                    │
                                    ▼
              ┌───────────────────────────────────────────┐
              │     ★ llm_tool_demo (This Module) ★       │
              │         Example Tool Implementations       │
              │  📖 6 Examples │ Best Practices │ Learn   │
              └─────────────────────┬─────────────────────┘
                                    │
                                    ▼
              ┌───────────────────────────────────────────┐
              │                   llm                     │
              │            (Core Base Module)             │
              └───────────────────────────────────────────┘

Installation

What to Install

For learning tool development:

odoo-bin -d your_db -i llm_tool_demo

Auto-Installed Dependencies

llm (core infrastructure)
llm_tool (tool framework)

What You'll Learn

Demo Tools

1. get_system_info

Location: models/res_users.py

Simple read-only tool to get Odoo system information.

@llm_tool(read_only_hint=True, idempotent_hint=True)
def get_system_info(self) -> dict:
    """Get basic Odoo system information..."""

2. calculate_business_days

Location: models/utility_tools.py

Utility tool with type hints for automatic schema generation.

3. create_lead_from_description

Location: models/crm_lead.py

Business logic tool that creates CRM leads (destructive operation).

4. generate_sales_report

Location: models/sale_order.py

Complex reporting tool with data aggregation.

5. get_record_info

Location: models/ir_model.py

Legacy code example with manual JSON schema.

6. send_notification_to_user

Location: models/res_users.py

User interaction tool for in-app notifications.

Key Concepts

Type Hints for Schema Generation

def my_tool(self, name: str, count: int = 10) -> dict:
    ...

Metadata Hints

read_only_hint=True - Doesn't modify data
idempotent_hint=True - Safe to call multiple times
destructive_hint=True - Modifies/creates/deletes data

Technical Specifications

Version: 18.0.1.0.0
License: LGPL-3
Dependencies: llm, llm_tool, crm, sale

License

LGPL-3

Please log in to comment on this module

The author can leave a single reply to each comment.
This section is meant to ask simple questions or leave a rating. Every report of a problem experienced while using the module should be addressed to the author directly (refer to the following point).
If you want to start a discussion with the author, please use the developer contact information. They can usually be found in the description.