The problem
Add placeholder for original text/plain MIME body
Would it be possible to add a separate placeholder that retrieves the original text/plain MIME part of an email directly from Thunderbird, when available?
Currently, with some Outlook-generated emails, {%mail_text_body%} appears to perform an HTML-to-text conversion. This can include Outlook/Word CSS (@font-face, MsoNormal, etc.) while also losing paragraph/line breaks.
For example, HTML containing:
DMS could be XXXX
Server 2TB
AMC 5*500 = 2500+Tax
can become:
@font-face ... MsoNormal ... DMS could be XXXXServer 2TBAMC 5*500 = 2500+Tax...
If Thunderbird exposes the original text/plain MIME part, a dedicated placeholder for it would be very useful for AI prompts.
Suggested behavior:
- Return the original
text/plain MIME content without HTML conversion.
- Keep it separate from
{%mail_text_body%} for backward compatibility.
- If no
text/plain MIME part exists, return empty (or optionally fall back to the existing conversion).
This would reduce token usage and provide cleaner, correctly structured email content, especially for Outlook emails.
Additional observation: If I switch Thunderbird to View → Message Body As → Plain Text and then use the existing {%mail_text_body%} placeholder, ThunderAI correctly extracts the clean, properly formatted plain-text version of the email.
Which Operative System are you using?
Windows 11
Which version of Thunderbird are you using?
154.0
Which version of ThunderAI has the issue?
4.1.1.0
Which integration are you using?
ChatGPT Web Interface
Anything in the Thunderbird console logs that might be useful?
Additional information
No response
The problem
Add placeholder for original
text/plainMIME bodyWould it be possible to add a separate placeholder that retrieves the original
text/plainMIME part of an email directly from Thunderbird, when available?Currently, with some Outlook-generated emails,
{%mail_text_body%}appears to perform an HTML-to-text conversion. This can include Outlook/Word CSS (@font-face,MsoNormal, etc.) while also losing paragraph/line breaks.For example, HTML containing:
can become:
If Thunderbird exposes the original
text/plainMIME part, a dedicated placeholder for it would be very useful for AI prompts.Suggested behavior:
text/plainMIME content without HTML conversion.{%mail_text_body%}for backward compatibility.text/plainMIME part exists, return empty (or optionally fall back to the existing conversion).This would reduce token usage and provide cleaner, correctly structured email content, especially for Outlook emails.
Additional observation: If I switch Thunderbird to View → Message Body As → Plain Text and then use the existing {%mail_text_body%} placeholder, ThunderAI correctly extracts the clean, properly formatted plain-text version of the email.
Which Operative System are you using?
Windows 11
Which version of Thunderbird are you using?
154.0
Which version of ThunderAI has the issue?
4.1.1.0
Which integration are you using?
ChatGPT Web Interface
Anything in the Thunderbird console logs that might be useful?
Additional information
No response