Prioritization & exploration

class Task:
    def __init__(self, name, priority, deadline=None):
        self.name = name
        self.priority = priority  # urgent/high/medium/low
        self.deadline = deadline

    def get_priority_score(self):
        scores = {"urgent": 4, "high": 3, "medium": 2, "low": 1}
        return scores.get(self.priority, 1)

class TaskPrioritizer:
    def prioritize(self, tasks):
        """Sort tasks by priority score."""
        return sorted(tasks, key=lambda t: t.get_priority_score(), reverse=True)

    def ai_prioritize(self, tasks, llm):
        """Use AI to analyze business impact and recommend order."""
        task_list = "\n".join([f"- {t.name} ({t.priority})" for t in tasks])
        prompt = f"Prioritize these tasks by business impact:\n{task_list}"
        return llm.generate(prompt).content

class ExplorationAgent:
    def __init__(self):
        self.llm = get_llm()
        self.knowledge = []

    def generate_hypotheses(self, topic):
        prompt = f"Generate 5 testable hypotheses about: {topic}"
        return self.llm.generate(prompt).content

    def explore_topic(self, topic, depth=2):
        hypotheses = self.generate_hypotheses(topic)
        for hypothesis in hypotheses:
            result = self.research_hypothesis(hypothesis)
            self.knowledge.append(result)
        if depth > 1:
            new_areas = self.discover_new_areas()
            for area in new_areas:
                self.explore_topic(area, depth - 1)

    def discover_new_areas(self):
        """Suggest unexplored areas based on accumulated knowledge."""
        prompt = f"""Based on these findings:
        {self.knowledge}
        What unexplored areas should we investigate next?"""
        return self.llm.generate(prompt).content