Memory
Workers remember what they learn, with the source, confidence and freshness of each memory. You can tell the workforce a fact, verify it, archive it or forget it.
What a memory hasLink to What a memory has
- a type: Working, Episodic, Semantic, Procedural or Social;
- a scope: the organization, a team, one worker or one work item;
- an authority: Explicit (stated in a document), Told (a person said it), Observed (measured from history) or Inferred (a model inferred it). Inferred memories always have lower confidence and never become policy;
- a status: Active, Superseded, Archived or Expired;
- a freshness, for example Fresh, Stale or Expires in some days.
Browse and recallLink to Browse and recall
Memory has two modes. Browse lists memories with filters. Recall asks the way a worker would, for example "How do we release?", and ranks memories by relevance, confidence, freshness and scope.
ActionsLink to Actions
- Members: Tell the workforce a fact, and Verify a memory.
- Admins: Archive (workers stop recalling it, but it is kept), Supersede, and Forget.
Forget removes the memory from what workers can find. Old audit entries and backups stay under their retention period.
Checked against the product on 2026-10-05.
Was this helpful?
Related articles
- DocumentsDocuments are sources that workers search and cite. Paste a document, or connect a source such as Notion or Confluence.
- Operating modelThe operating model holds the conventions Agent learned from your history, such as how you review and release. Nothing is used until a person confirms it.
- LessonsWhen work finishes and people give feedback, workers propose lessons. An approver accepts or rejects each one before it is used.