ArchitectureA supervised JVM-class runtime — OLTP on seven engines, OLAP on three. AI-native, MCP-native, observable as plain SQL.Read the architecture
Está viendo la edición Perú. Está viendo la edición Colombia. You're viewing the Pakistan edition. Cambiar a la edición global →Cambiar a la edición global →Switch to the global edition →

Orchestrator and sub-agents share a conversation-scoped artifact scratchpad

A key/value store propagated across the orchestrator and all its sub-agents lets each agent read artifacts a sibling already produced, eliminating redundant re-fetches when the same data is needed across delegation boundaries.

When an agent delegates to a sub-agent, the sub-agent runs in an isolated context: its intermediate results live only in its own memory and are discarded on return. A follow-up delegation that needs the same data has to re-fetch it from source. A conversation-scoped artifact scratchpad eliminates that redundancy by propagating a shared store to the orchestrator and every sub-agent it spawns.

How the scratchpad works

  • Artifact capture. A tool result flagged as reusable is captured into the scratchpad automatically when the agent loop processes it — no extra round-trip to the language model. Entries are bounded by a configurable count ceiling.
  • Shared by reference. The same store instance is propagated to every sub-agent context. A sub-agent receives a key index at delegation time so it can read prior artifacts with scratchpad_read without knowing whether the orchestrator or a sibling produced them.
  • Explicit pinning. An agent can write a named entry with scratchpad_write to make it permanently visible to the rest of the session; auto-captured entries are managed by the weigher, pinned entries are not evicted.

Why it matters for multi-agent deployments

  • Fewer redundant tool calls. Data that one sub-agent fetched — a query result, a resolved lookup, a computed summary — is available to every subsequent delegation without hitting the source again.
  • Bounded memory. The store is Caffeine-backed with a configurable entry limit, so a long-running session does not accumulate unbounded state.

The scratchpad is opt-in per agent configuration and integrates with the existing delegation model. Agents that do not load the scratchpad toolkit continue to operate unchanged.

See the feature →

← All posts