Scoring line: colon restored, full sentence (score over N samples), score keeps solid green with yellow phrase tail; path coloring covers 'to /abs' endings

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

evalharness/cli.py

@@ -228,10 +228,16 @@ def _narration(msg: str) -> str:
     # gets bold green -- it is what the eye should find first
     import re as _re1
 
-    m = _re1.search(r'[:·] ([a-zA-Z_@]+ [0-9.]+%)$', msg)
+    m = _re1.search(r'[:·] ([a-zA-Z_@]+ [0-9.]+%)(?=\s|$)', msg)
     if m:
-        return (f'{icon}{msg[:m.start()]}'
-                f'[bold green]{m.group(1)}[/bold green]')
+        head = f'{icon}{msg[:m.start()]}: [bold green]{m.group(1)}[/bold green]'
+        tail = msg[m.end():]
+        if tail:
+            import re as _re2
+
+            tail = _re2.sub(r'(?<![\w/%.])(\d+(?:/\d+)?(?:\s+[a-z-]+){0,4})(?=[\s,.]|$)',
+                            r'[bold yellow]\1[/bold yellow]', tail)
+        return head + tail
 
     # semantic PHRASES get color, not bare numbers: '164 samples from cache',
     # '4/4 predictions' -- a lone number is easy to lose. Runs AFTER the
@@ -240,10 +246,11 @@ def _narration(msg: str) -> str:
 
     msg = _re0.sub(r'(?<![\w/%.])(\d+(?:/\d+)?(?:\s+[a-z-]+){0,4})(?=[\s,.]|$)',
                   r'[bold yellow]\1[/bold yellow]', msg)
-    # highlight any trailing path after '->'
-    if '-> ' in msg:
-        head, _, tail = msg.partition('-> ')
-        return f'{icon}{head}-> [cyan]{tail}[/cyan]'
+    # highlight any trailing path after '->' or 'to' (only absolute paths)
+    for sep in ('-> ', 'to '):
+        head, _, tail = msg.rpartition(sep)
+        if head and tail.startswith('/'):
+            return f'{icon}{head}{sep}[cyan]{tail}[/cyan]'
     return f'{icon}{msg}'
 
 

evalharness/model/runner.py

@@ -747,7 +747,8 @@ async def run_eval(
         _m = next(((k, v) for k, v in report.metrics.items()
                    if k != 'extraction_failure_rate'), None)
         status_callback(f'Scoring complete: {_m[0]} {_m[1] * 100:.1f}% '
-                        if _m else 'scoring complete')
+                        f'over {report.num_samples} samples'
+                        if _m else 'Scoring complete')
     # performance profile: pool success rate + latency/ttft percentiles
     try:
         from .aggregator import get_aggregator