His agent keeps its own handover notebook; he just says 'remember that' after each mistake
The operator of a small Japanese company has used AI daily for three months across quote tooling, documents, internal pages and accounting help, and never edits a file himself: after every failure he tells the agent in chat to remember the lesson, and the agent writes it into its own handover note and memory (now 250+ memory entries and a note past 1,000 lines). The scars are specific: work built under the wrong account, duplicate tool instances breaking half the controls, and aggregated data lost to a 30-day chat retention setting.
How it works
- Owner asks for work in plain chat: quotes, documents, internal pages, accounting help
- When something fails, he says 'remember that, and be careful next time'
- Agent writes the lesson, with its reason and date, into its handover note or memory itself
- Each new session starts by reading the note
- Memory passes 250 entries in three months and is re-indexed by topic at his request
- His public tools site is built entirely through these conversations
Why it’s worth knowing
The memory is maintained entirely by the agent: the human's whole configuration interface is one spoken sentence after each failure, and the notebook still grew past a thousand lines.
Where it’s from
Summarized from a public post on note.com. See the original ↗