I've been trying to solve a problem of scattered operational context across Slack threads, tickets, meetings, etc. (my way)... The idea of Trailogs is to be a central event-based history, where each event (log) has structure around it, like categorization, ownership, etc...
It has an exploration page, which is basically an AI chat where you can ask questions about what is happening in the company, why a certain decision was made, what is going on with a specific customer, who was responsible for a certain thing, etc...
I realized that "what should I ask?" could be a problem, so the chat gives you suggestions based on your role in the team and interests. So if you are a developer, you could get a suggestion to ask about a certain deployment/feature release from today. If you are in sales, you'll get a suggestion to ask about negotiations that happened today with a certain customer.
The chat is currently powered by OpenAI. Now, I understand the privacy concerns and sensitivity of the data, so I'm planning an implementation with a self-hosted LLM, or the possibility for users to use their own LLM.
I was thinking about how to reduce the friction so teams don't have to manually log stuff every time, so, for now, I implemented the possibility to integrate through an API and webhooks (incoming, outgoing). Also, since a lot of team conversations happen in Slack, I've built a Slack integration so you can ask questions directly from Slack or ask it to draft a log based on what was discussed in a thread.
The app is functional and can be tried. It has a registration form, but I made it possible to just continue with a demo account, so you don't have to leave any personal information if you don't want to, and you can quickly get straight to the demo workspace.
Poke around, let me know what you think. :)