Engine layer
Close the loop: predict, act, record, recalibrate.
Most memory layers stop at retrieval. They hand your agent some context and that is the end of it. Flows is the half that does something with it. New facts and forecasts act as triggers, steps call your systems, low-confidence decisions route to a person, and the result is written back into memory. That result is what corrects the next forecast.
flows · triggers · runs · learning.recordOutcome
Features
Memory and forecasts as triggers
Fire on a new fact, a changed belief, a threshold crossing, or a forecast clearing a confidence bar. Not just webhooks and cron.
People in the loop
Send low-confidence decisions to a person. Their call is recorded as an outcome, so the engine learns from it.
Outcomes write back
Every run records what was decided and what happened. That feeds calibration instead of vanishing into a log file.
Visual builder or code
Build flows in the wizard, or drive the same engine from the SDKs and MCP.
More engine layers
Memory
Recall that compounds.
Cross-session persistent memory with hybrid search and a knowledge graph. The engine decides what's worth keeping, so context carries between conversations without bloating your prompts.
memory.observe · memory.search · memory.reflect
Predictions
Foresight, not just recall.
Declare any target and the engine predicts it from a subject's history — calibrated, provenanced, and abstaining when the signal is thin. This is the layer mem0 and MemoryLake have no answer for.
lattice.predict (event · numeric · time · anomaly)
Behaviors
From patterns to next-best-action.
The engine mines behavior into emergent patterns and closes the loop: predict, let the agent act, record the outcome, and watch the patterns recalibrate. Decisions and outcomes are first-class.
lattice.mine · learning.recordDecision / recordOutcome
Intelligence that acts, not just answers.
An automation engine built on the memory graph. A forecast fires or a new fact lands, and a flow calls a system, runs a workflow, brings in a person, then writes the result back.