An agent memory hierarchy keeps frequently needed working state near the context window and durable history and source material external until retrieval justifies their token cost. Tiers differ by scope, function, lifetime, access, and authority; proximity changes availability, not authority. section 4.1.1
Table 4.1 Dimensions of agent memory
|
Memory category
|
Example content
|
Scope
|
Function
|
Lifetime and access
|
Owner or authority
|
|
Active context
|
Current instructions, evidence, tool results
|
Turn or session
|
Working state
|
Immediate and volatile
|
Runtime plus source owners
|
|
Task or session state
|
Goal, artifacts, receipts, pending decisions
|
Task or session
|
Resumption
|
Durable until task closure
|
Orchestrator or task owner
|
|
Project or user context
|
Conventions, preferences, profiles
|
Project, user, or team
|
Continuity
|
Durable, versioned, retrieved on demand
|
Human or system owner
|
|
Admitted lessons
|
Scoped guidance with admission evidence
|
Workload, team, or organization
|
Governed reuse
|
Versioned, expiring, supersedable
|
Admission owner
|
|
Archival evidence
|
Raw traces, prior versions, audit records
|
Task through organization
|
Audit and later review
|
Archived under retention policy
|
Evidence or records owner
|
Compare memory categories by scope, function, lifetime, access, and owner or authority. section 4.1.1
Keep placement separate from correctness and control:
- Frequency and recency do not establish correctness; review model-written notes and set precedence for conflicting versions. section 4.2.6
- Before a high-risk decision, retrieve the authoritative control-plane source and record which version governed the action. section 4.2.6




