Shared does not have to mean cloud-owned
The useful part of memory is often the sensitive part: decisions, incidents, customer constraints, internal workflows, and exact commands.
Shared AI memory
Shared AI memory that stays under your control
shared
Palace Server lets engineering teams share agent recall without giving up control of the database, embeddings, or deployment boundary.
Runs on your infrastructure, not Palace's cloud.
Local ONNX embeddings: no third-party embedding API required.
Ed25519-signed licenses verify offline for air-gapped operation.
What gets shared
Palace is tuned for coding-agent memories: diaries, preferences, source-grounded recall, KG facts, project wings, and remembered fixes.
How teams adopt
Start with the free local core, then deploy Palace Server when shared recall becomes valuable enough to centralize.