How CLS++ Memory Works
A simple, powerful system for maintaining conversation context across any LLM provider.
User Chats with Any LLM
Your user starts a conversation with Claude, GPT, Gemini, or any supported LLM. The conversation flows naturally.
Context Stored via CLS++
After each exchange, CLS++ captures and stores the conversation context, including preferences, decisions, and key information.
User Switches Providers
When your user switches to a different LLM (GPT, Gemini, etc.), they simply continue the conversation.
Context Auto-Retrieved
CLS++ automatically retrieves and formats the context for the new provider. The conversation continues seamlessly.
Built for Scale
Reconsolidation gate
Every revision is evidence-gated: a contradicting fact archives the old one with lineage and version. Memory is updated — never silently overwritten or deleted.
Durable write path
Writes are fsync-acknowledged to a write-ahead log with off-path replication. Recovery is tested against process death and torn-tail records — an accepted write is a kept write.
Embedder-independent retrieval
Facts persist and stay searchable from the engine's own token index, so a failed or slow embedder degrades ranking but never loses a memory.
Yours to control
Encrypted in transit; delete any memory on request (GDPR right to be forgotten). We don't claim compliance badges we don't hold.
Data Flow
See it live
Type one fact, watch three models recall it — then run the reliability proof to see it never silently loses or overwrites it.