migrate-memory

Migrating agent memory when switching providers requires transferring durable facts and identity instructions between distinct storage systems. This skill automates the synchronization, ensuring the process is idempotent and non-destructive to prevent data loss.

Transferring agent memory from a flat file to a scaffolded directory is best handled by automated cross-architecture translation. This skill reads flat file structures and translates them into scaffolded directory trees without manual copy-pasting.

Yes, memory migration filters out transient conversational residue. Through intelligent distillation, the process retains only core agent identity and user-defined facts while discarding temporary chat history during the synchronization.

Yes, you can sync memory stores without data loss because the migration process is designed to be idempotent and non-destructive. It safely transfers durable facts and identity instructions across different agent infrastructure configurations.

You need to migrate memory between agent infrastructure configurations when switching from one AI provider to another, such as moving from a Claude-based provider to a Codex-based provider, to preserve the agent's personality and historical context.