As PDW, I want the AI assistant to track all tasks in my organization and group them by project, so I can ask questions and receive answers informed by my organizational data

[n0fray]

• [ ] When user is chatting with Tachio, if a to-do item has been described that does not exist, a new item should be created
• [ ] This new task should be associated with the related project
• [ ] Project comes from the context of the conversational thread
• [ ] If conversation thread is not clear, and project association cannot be determined then Tachio should ask for clarification
• [ ] Tachio understands due date from conversational content
• [ ] If Tachio does not understand due date from conversational content, then should ask for clarification on due data
• [ ] If due date is given, should be connected to the associated task and scheduled accordingly

[n0fray]

Here is a list of the attributes that we will want the ability to query on:

• ID
• Todo name
• Description
• Status (already done, in progress, todo)
• Priority (Now, next, later)
• Project ID
• Project’s Missive conversation id
• Partner organization id
• Related people/contacts array of ids
• External URLs (e.g., linear or github issue or pull request)
• Attachments (URLs of files in Supabase storage)
• Who created this todo, or if Tachio created this todo on their own, then what is the reason that this todo was created
• Which missive conversations are associated with this todo, array of conversation ids
• Date-time when this todo was created
• Date-time we started work on this todo
• Date-time we completed work on this todo
• Who is this todo assigned to (default is null, which means the whole team shares responsibility)
• Parent, child, peer relationships to any other todo(s) that are related to this one
• Vector embeddings of the todo name+description

[ejfox]

This is stale, we will not be adding all of these attributes to our todos table