Principle
Separate session state from long-term memory
What the agent needs for the next hour is not the same thing it should remember next month.
When it fits
- A project spans multiple conversations, workers or models.
When to avoid it
- Some transient-looking events become important evidence after failure; retain raw logs according to audit and incident needs even if they are not promoted into memory.
Why it matters
Keep short-lived execution state—current step, temporary errors, open tool handles—separate from durable project knowledge such as decisions, accepted terminology, architecture, source references and unresolved risks. Persist only the latter deliberately. This makes it possible to restart a worker without turning every transient event into permanent memory.
An example
A failed test command belongs in the session log; the discovery that a specific interface requires a legacy mapping belongs in durable project memory.
Check your result
A new worker can recover durable decisions without inheriting every temporary detour.
Keep this limit in mind
- Some transient-looking events become important evidence after failure; retain raw logs according to audit and incident needs even if they are not promoted into memory.
Connected ideas
Use beforeGive agent memory a write-manage-read policy
Evidence and sources
Current long-horizon agent practice distinguishes transient working context from durable state that can survive context resets or worker replacement.
A durable log is one architecture; structured snapshots or databases may be better for other systems.
The Log Is The Agent · Durable history, disposable execution