Add safety checks that catch risky prompts before AI responds

flowchart TD trig>"customer message received"]:::trig s0["screen incoming message"]:::task s1["generate ai response"]:::svc s2["screen ai response"]:::task s3["deliver final reply"]:::task trig --> s0 s0 --> s1 s1 --> s2 gx{"× is content flagged"}:::gate s2 --> gx p00["send pre written fallback reply"]:::task gx -->|"flagged content"| p00 p10["send ai generated reply"]:::task gx -->|"content is safe"| p10 jn{"○ reply sent"}:::gate p00 --> jn p10 --> jn jn --> s3 out[/"safely filtered ai reply delivered"/]:::out pay{{"lower risk of harmful ai replies"}}:::pay s3 --> out out --> pay s2 -. "failures recorded, run continues" .-> out classDef task fill:#e7f6fe,stroke:#34b8f0,color:#2c2a29 classDef svc fill:#f6f8fa,stroke:#7c8795,color:#2c2a29 classDef mi fill:#e7f6fe,stroke:#0079a8,color:#2c2a29,stroke-width:2px classDef human fill:#fff,stroke:#0079a8,color:#0079a8 classDef store fill:#f6f8fa,stroke:#0079a8,color:#2c2a29 classDef trig fill:#00a4eb,stroke:#0079a8,color:#fff,font-weight:bold classDef trigtime fill:#00a4eb,stroke:#0079a8,color:#fff,font-weight:bold classDef trigdata fill:#8ad4f5,stroke:#0079a8,color:#06314c,font-weight:bold classDef gate fill:#fff,stroke:#e8a23d,color:#6b4708,font-weight:bold classDef out fill:#1f9d6b,stroke:#167a53,color:#fff,font-weight:bold classDef pay fill:#06314c,stroke:#021f33,color:#fff