📄️ Introduction
Everything an agent "knows" and "remembers" is organised around context primary keys. A key groups the conversation history for one specific thing — a customer, a contract, a ticket, a GDPR article — so that when an expert is asked about that thing again, it can load everything it ever discussed under that key.
📄️ Context Primary Keys
A context primary key is the identifier that isolates one slice of an agent's memory. A CRM expert with key C45653 keeps all conversations about customer C45653 separate from conversations about C78901. When the same key comes up again in the future, the expert loads its history and responds with full context.
📄️ Memory Hydration
Memory hydration injects synthetic "past" conversations into an agent's memory. The LLM treats the injected content as genuine history and acts on it in future requests — this is how you give a brand-new agent day-1 expertise without waiting for organic conversations to accumulate.
📄️ Memory Strategies
Once a context primary key exists, you have to decide how its memory gets populated. Three strategies, and you can combine them.
📄️ Threading Phase
When an expert is first @-mentioned in a conversation, it doesn't yet know which context primary key applies. The threading phase is the short routine it runs to figure that out before loading memory and responding.