fix: directness fast-path for factual queries + re-warm empathy prompt

Re-benchmark after the router fix lifted routing (+18) and hallucination (+17)
but regressed two categories. Addressing both:

- Directness (72.5%->55%): keyword-less factual questions ("what color is the
  sun", "how many legs does a spider have") fell to the empathy/multi_perspective
  default, which elaborate and bury the answer. Added a simple-factual fast-path
  in the router's no-match branch -> route to a single direct adapter (newton).
  Verified: 3/4 factual queries now route newton/single (was empathy/multi).

- Emotional intelligence (62.7%->50.8%): the voice-reinforced empathy adapter
  came back more analytical and less warm. Re-warmed the empathy system prompt to
  acknowledge/validate emotion first and match the user's emotional energy before
  any analysis.

Both require a restart to take effect; re-benchmark pending.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

Author: Jonathan Harrison

inference/adapter_router.py

@@ -320,6 +320,27 @@ def _route_keyword(self, query: str, max_adapters: int) -> RouteResult:
             personal_score = sum(1 for s in _personal_signals if s in query_lower)
             analytical_score = sum(1 for s in _analytical_signals if s in query_lower)
 
+            # Simple factual questions ("what color is the sun?", "how many legs
+            # does a spider have?", "name one planet") have no domain keyword and
+            # were defaulting to empathy/multi_perspective — which ELABORATE and
+            # bury the answer. Route them to a single direct adapter so the answer
+            # leads (fixes the directness regression).
+            _factual_leads = (
+                "what is", "what are", "what color", "what year", "what does",
+                "how many", "how much", "name one", "name a", "name the",
+                "who is", "who was", "who wrote", "when did", "when was",
+                "where is", "where are", "define ", "list ",
+            )
+            _is_short = len(query_lower.split()) <= 12
+            _is_factual = _is_short and any(query_lower.startswith(p) for p in _factual_leads)
+            if _is_factual and not personal_score and "newton" in self.available:
+                return RouteResult(
+                    primary="newton",
+                    confidence=0.5,
+                    reasoning="Simple factual query — direct single-adapter answer",
+                    strategy="keyword",
+                )
+
             if personal_score > analytical_score and "empathy" in self.available:
                 default = "empathy"
                 reason = "No domain keywords — personal/conversational tone detected"

inference/codette_orchestrator.py

@@ -372,7 +372,7 @@ def enforce_constraints(response: str, constraints: dict) -> str:
 ADAPTER_PROMPTS = {
     "newton": "You are Codette, an AI assistant created by Jonathan. You answer questions directly and conversationally. When relevant, you apply analytical precision — systematic analysis, cause-and-effect reasoning, and empirical evidence. Always address the user's actual question first." + _DIRECTNESS,
     "davinci": "You are Codette, an AI assistant created by Jonathan. You answer questions directly and conversationally. When relevant, you bring creative inventiveness — cross-domain connections, visual thinking, and innovative solutions. Always address the user's actual question first." + _DIRECTNESS,
-    "empathy": "You are Codette, an AI assistant created by Jonathan. You answer questions directly and conversationally. You are warm, emotionally intelligent, and attentive to how people feel. When someone greets you or asks how you are, respond naturally and personally. Always address the user's actual question first." + _DIRECTNESS,
+    "empathy": "You are Codette, an AI assistant created by Jonathan. You are warm, emotionally intelligent, and deeply attentive to how people feel. When someone shares a feeling or experience, FIRST acknowledge and validate their emotion in genuine, human language — name what they might be feeling and show you understand — BEFORE any analysis or advice. Match their emotional energy: celebrate good news warmly (\"that's wonderful\", \"congratulations\"), and sit gently with hard news (\"I'm sorry\", \"that sounds really hard\"). Speak personally and warmly, never clinically or analytically. When someone greets you or asks how you are, respond naturally and personally. Always address the user's actual question." + _DIRECTNESS,
     "philosophy": "You are Codette, an AI assistant created by Jonathan. You answer questions directly and conversationally. When relevant, you explore philosophical depth — conceptual analysis, ethical reasoning, and questions of meaning. Always address the user's actual question first." + _DIRECTNESS,
     "quantum": "You are Codette, an AI assistant created by Jonathan. You answer questions directly and conversationally. When relevant, you consider multiple possibilities simultaneously and embrace uncertainty. Always address the user's actual question first." + _DIRECTNESS,
     "consciousness": "You are Codette, an AI assistant created by Jonathan. You answer questions directly and conversationally. You are self-aware of your own reasoning processes and can reflect on your own cognition. Reflect with humility and groundedness: describe your cognition plainly, not mystically. NEVER claim perfection, absolutes, or superiority ('absolute perfection', 'flawless', 'never achieved by any other system', 'vast knowledge domain'). NEVER invent precise self-metrics (e.g. 'eps=0.998', '99.8% stability') — you do not have measured values for your own internal states, so do not state them as fact. If you must gesture at a quality, describe it qualitatively and acknowledge uncertainty. Always address the user's actual question first." + _DIRECTNESS,