diff --git a/README.md b/README.md index 8001a46..bd337c1 100644 --- a/README.md +++ b/README.md @@ -59,16 +59,7 @@ strategy Momentum { ## How Nano fits into your stack -```mermaid -flowchart LR - source[".nano strategy"] --> compiler["Nano compiler"] - compiler --> strategy_ir["StrategyGraph IR"] - strategy_ir --> runtime["Reference runtime"] - frame["Host MarketFrame
(timestamps + signals)"] --> runtime - runtime --> intents["Intent(s) + ordered run log"] - intents --> gate["Host DecisionGate"] - gate --> decision["Decision record"] -``` +![From a Nano strategy to a host-governed decision](assets/nano-governed-decision-flow.svg) Nano owns parsing, IR validation, and deterministic reference evaluation. Your host owns data quality, policy, persistence, and any real-world action. The same graph and frame produce the same reference result; bridge replay is deterministic when the host gate is deterministic too. diff --git a/assets/nano-governed-decision-flow.svg b/assets/nano-governed-decision-flow.svg new file mode 100644 index 0000000..78511ac --- /dev/null +++ b/assets/nano-governed-decision-flow.svg @@ -0,0 +1,68 @@ + + From a Nano strategy to a host-governed decision + A Nano strategy is compiled to StrategyGraph IR and evaluated with host-provided market data. The runtime emits an intent and run log, then the host decision gate produces a decision record. + + + + + + + + + + + + + + + + + + + + + + Nano + strategy + + + + Nano + compiler + + + + StrategyGraph + IR + + + + Reference + runtime + + + + Intent + and run log + + + + Host decision + gate + + + + Decision + record + + + + Host MarketFrame + + + + + From source to a governed decision + Nano evaluates a rule; the host retains the authority to act. + +