7 Boundaries and Invalid Claims · 7.4 Not Chatbot Wrapper
7.4 Not Chatbot Wrapper
Human Behavior FCA
This page is structured as definition, control variables, causal chain, observable outputs, and boundary, and serves as a canonical definition node in Human Behavior FCA.
FCA is not a chatbot wrapper and not a prompt wrapper; it is a behavior-generation architecture.
A chat interface is only an input/output carrier and not the system itself.
System value comes from long-term state tracking, breakpoint localization, and Feedback Update rather than reply style.
Whether long-term cross-cycle state is recorded.
Whether Observable Output is tracked beyond chat text.
Whether SN, DMN, FP, ECN, and Feedback nodes are localized.
Whether Feedback Update is recorded and written back to state.
Whether one user keeps chain continuity across cycles.
Observable Output → Foreground Entry → Historical Template Matching and Explanation → New Structure Generation → Execution Maintenance → Feedback Update.
Chat input is only an input carrier and cannot replace the behavior-generation chain.
If the system outputs answers only without chain-state maintenance and feedback write-back, it is not an FCA runtime.
Reply style changes but state continuity is missing.
The system performs QA only and does not store feedback updates.
It cannot distinguish FP absence from ECN breakdown.
The next user cycle starts from zero without historical chain carryover.
Do not treat UI as runtime.
Do not treat prompts as architecture.
Do not treat one-shot chat as a long-term state system.
Do not expose backend scoring and parameter details.