New: Cookbooks and AI ExplanationsStep-by-Step recipes to solve problems connected to Roadmaps and Cheat Sheets. Need more details? Use AI buttons for structured and simple explanations with concrete examples throughout the whole platform.Take a look
Your coding agent relearns the same project quirks and past decisions from scratch every single run.
What you'll have at the end
A memory file the agent updates after each run and reads first at the start of the next one
You need
An agent that already runs multi-step tasks on your own project across more than one separate run, with a place on disk it can read and write plain text files.
Not covered
Retrieving semantically similar memories with embeddings, or sharing one memory across several different agents; this recipe covers one agent's own file, read and written directly by that agent alone.
Leans on
Should your agent ask before it deletes, sends, or pays?
If the risky move a memory line might drive is one that deletes, sends, or pays, decide whether it needs a person's go-ahead before it fires, not only a recheck against stored memory.
Log every tool call so you can find what broke a run
If what you need is a full record of every tool call for debugging one run, not a curated file of lessons that survive across runs, build that log instead.
Checked 18 Aug 2026
Part of the AI Agents cookbook