feat(eval): improve eval runner with structured output

Refactor the eval runner for cleaner result reporting and
better error handling across evaluation scenarios.

eval/cmd/eval-runner/main.go

@@ -78,19 +78,35 @@ func readPrompt() (string, error) {
 		return os.Args[2], nil
 	}
 
+	// promptfoo exec: provider passes the prompt as the first positional argument
+	if len(os.Args) > 1 && !strings.HasPrefix(os.Args[1], "-") {
+		raw := os.Args[1]
+		var promptData struct {
+			Prompt string `json:"prompt"`
+		}
+		if json.Unmarshal([]byte(raw), &promptData) == nil && promptData.Prompt != "" {
+			return promptData.Prompt, nil
+		}
+		return strings.TrimSpace(raw), nil
+	}
+
+	// Fallback: read from stdin (for manual testing / piping)
+	stat, _ := os.Stdin.Stat()
+	if (stat.Mode() & os.ModeCharDevice) == 0 {
 		data, err := io.ReadAll(os.Stdin)
 		if err != nil {
 			return "", err
 		}
-
 		var promptData struct {
 			Prompt string `json:"prompt"`
 		}
 		if json.Unmarshal(data, &promptData) == nil && promptData.Prompt != "" {
 			return promptData.Prompt, nil
 		}
-
 		return strings.TrimSpace(string(data)), nil
+	}
+
+	return "", fmt.Errorf("no prompt provided (use positional arg, --prompt, or pipe to stdin)")
 }
 
 func loadEvalConfig() (*config.Config, error) {
@@ -217,7 +233,6 @@ func emitError(msg string) {
 	trace := Trace{Error: msg}
 	out, _ := json.Marshal(trace)
 	fmt.Println(string(out))
-	os.Exit(1)
 }
 
 type instrumentedCall struct {