Componenti aggiuntivi
Benvenuto in Thunderbird Add-ons.
Aggiungi nuove funzionalità e stili per rendere Thunderbird veramente tuo.
Chiudi
Informazioni su questo componente aggiuntivo
About assistant
This Thunderbird Extension provides a powerful and flexible framework for AI-powered email analysis. It automatically processes incoming emails, sends them to a language model of your choice for analysis, and then applies tags based on the model's response. This allows for sophisticated, automated email classification and sorting.
How to use
Configuration and setup
After installing the add-on, you can configure it by going to Tools > Add-ons and Themes, finding "Mail Assistant", and clicking the "..." button to select Options.
The General tab allows you to select your preferred LLM provider and enter the necessary credentials (API Key or local URL).
The Custom Tags tab is where you can define the categories for email analysis. You can add, edit, or delete tags. Each tag has a name, a unique key, a color, and a specific prompt instruction that tells the LLM what to check for.
Once configured, the extension will automatically process new incoming emails. It applies the tags generated by the AI analysis, which can then be used with Thunderbird's built-in Message Filters to organize your inbox.
Organizing emails
Once automated tagging is working, you may create filters that will move tagged messages to folders or take other actions. You should set those filters to run periodically, as tagging is executed after initial filtering.
Privacy disclaimer
If you are using the LLM engine other than the local Ollama, all your emails will be sent to the external provider to process, which may also include sensitive information. Make sure you understand how your LLM provider processes your data.
This Thunderbird Extension provides a powerful and flexible framework for AI-powered email analysis. It automatically processes incoming emails, sends them to a language model of your choice for analysis, and then applies tags based on the model's response. This allows for sophisticated, automated email classification and sorting.
- Multi-Provider LLM Support: Integrates with local models via Ollama and cloud-based models from OpenAI, Google Gemini, Anthropic Claude, Mistral, and DeepSeek.
- Dynamic Email Analysis: Intelligently extracts headers, text content (converting HTML to plain text), and attachment details for efficient and accurate analysis by the LLM.
- Fully Configurable Tagging: Allows users to define their own custom tags, colors, and LLM prompts for a completely personalized email classification system.
How to use
Configuration and setup
After installing the add-on, you can configure it by going to Tools > Add-ons and Themes, finding "Mail Assistant", and clicking the "..." button to select Options.
The General tab allows you to select your preferred LLM provider and enter the necessary credentials (API Key or local URL).
The Custom Tags tab is where you can define the categories for email analysis. You can add, edit, or delete tags. Each tag has a name, a unique key, a color, and a specific prompt instruction that tells the LLM what to check for.
Once configured, the extension will automatically process new incoming emails. It applies the tags generated by the AI analysis, which can then be used with Thunderbird's built-in Message Filters to organize your inbox.
Organizing emails
Once automated tagging is working, you may create filters that will move tagged messages to folders or take other actions. You should set those filters to run periodically, as tagging is executed after initial filtering.
Privacy disclaimer
If you are using the LLM engine other than the local Ollama, all your emails will be sent to the external provider to process, which may also include sensitive information. Make sure you understand how your LLM provider processes your data.
