Working memory is the information available in an agent’s context window during a model call: fast, small, and volatile. It can include current instructions, evidence, and tool results; finite token capacity makes context placement an architectural choice. section 4.1.1 section 1.1
Keep frequently needed state near the model; leave durable history and source material outside the context until retrieval justifies their token cost. Moving information closer changes its availability, not its authority: provenance, version, ownership, and invalidation still matter. 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 active context with task state, project or user context, admitted lessons, and archival evidence by scope, function, lifetime, and authority. section 4.1.1










