The Kapso feedback generator and the decision to stop

@dataclass
class FeedbackResult:
    stop: bool                      # Whether to stop iteration
    evaluation_valid: bool          # Whether evaluation is fair/correct
    feedback: str                   # Actionable feedback for next iteration
    score: Optional[float] = None   # Extracted evaluation score
    cards_load_bearing: List[str]   # Served bank cards whose guidance mattered
    cost_usd: float = 0.0           # Cost of the feedback agent call
    duration_seconds: Optional[float] = None

# Example evaluation output
"Training complete. Accuracy: 0.95, F1: 0.92"

# Extracted score
score = 0.95  # Primary metric

feedback = """
Current accuracy is 0.85, target is 0.9.
Suggestions:
1. Try increasing the number of estimators in the random forest
2. Consider feature engineering to improve signal
3. Check for class imbalance and apply SMOTE if needed
"""

solution = kapso.evolve(
    goal="Build a classifier with accuracy > 0.9",
    max_iterations=10,
)

# Check if goal was achieved
if solution.succeeded:
    print(f"Goal achieved with score: {solution.final_score}")
else:
    print(f"Final feedback: {solution.final_feedback.feedback}")

from kapso.execution.search_strategies.generic import FeedbackGenerator, FeedbackResult

# Create feedback generator (uses claude_code by default)
feedback_gen = FeedbackGenerator()

# Generate feedback — the judge reads the diff itself from the workspace
result = feedback_gen.generate(
    goal="Build a classifier with accuracy > 0.9",
    idea="Use random forest with hyperparameter tuning",
    code_changes_summary="Implemented RF with GridSearchCV",
    base_branch="main",
    head_branch="generic_exp_3",
    evaluation_script_path="kapso_evaluation/evaluate.py",
    evaluation_result="Accuracy: 0.85",
    workspace_dir="./iris_solution",
)

if result.stop:
    print("Goal achieved!")
else:
    print(f"Continue with feedback: {result.feedback}")

feedback_generator:
  type: "claude_code"
  model: "claude-opus-5"
  agent_specific:
    auth_mode: "oauth"

# In the generic strategy's iteration
solution = generate_solution(context)          # ideation
node = run_implementation(solution, branch)    # coding agent implements + evaluates
feedback = generate_feedback(node)             # the judge reads the session's tags

node.feedback = feedback.feedback
node.should_stop = feedback.stop
# The judged score stands only if the evaluation was ruled valid;
# an invalid evaluation nulls it, and the run's manifest-of-record
# score overrides the judge's reading when present.
node.score = feedback.score if feedback.evaluation_valid else None

# Good: Clear criteria
goal = "Build a classifier with accuracy > 0.9 and F1 > 0.85"

# Less good: Vague criteria
goal = "Build a good classifier"

# Agent builds evaluation dynamically
solution = kapso.evolve(
    goal="Optimize inference speed to < 10ms per sample",
    max_iterations=10,
)

Iteration 1: Accuracy 0.75
Feedback: "Try ensemble methods and feature selection"

Iteration 2: Accuracy 0.88
Feedback: "Close to target. Try hyperparameter tuning"

Iteration 3: Accuracy 0.92
Feedback: "Goal achieved! Accuracy 0.92 > 0.9 target"