Note · 2 min read
The context that turns a chat into an assistant
Hermes saves its activity in Obsidian through a daily cron and links between notes. That memory lets me return to earlier topics and work with accumulated meeting, task and product context.
Recover work from months ago
Hermes uses Obsidian as persistent memory. A daily cron records everything it has done that day and links the notes. This record lets it retrieve earlier work and connect topics across conversations. I can return to any subject we have worked on without having to explain it again.
If I speak to a lead in March, I want to ask Hermes in June why they did not buy and what happened in the meeting. A scheduled job collects transcripts; the CRM and lead research provide the rest of the context. That also helps prepare new meetings and adapt outbound messages.
I use the same assistant to review tasks and ideas I write in Notion and decide what to work on next. It now also has read-only access to selected Invisor tables and reads Instagram and TikTok statistics.
Use the context to make decisions
Its weekly report found that users got stuck after writing their strategy. They left without analysing a property; some activated a plan before trying the analysis.
I changed that part of onboarding. Before the change, there had been two analyses in total since the PLG launch. Afterwards, seven in one day, with the same visits and logins.
The assistant’s usefulness depends on the information it can retrieve and connect. Transcripts let it remember meetings; tasks help it support prioritisation; usage data helps it identify product problems. Managing that context expands the work I can delegate and reduces what I have to reconstruct manually in each conversation.