fix: clamp retry_attempts in CriticAgent like ChatAgent

[aniketkrs]

User impact

CriticAgent(retry_attempts=0) never consults the model: get_option() skips its retry loop entirely and silently returns a random option. Users passing retry_attempts=0 get an ungrounded answer with no indication that the LLM was never called.

Fixes #4378.

Technical cause

ChatAgent.__init__ clamps the value (chat_agent.py:649, self.retry_attempts = max(1, retry_attempts)), but CriticAgent.__init__ overwrites it with the raw argument (critic_agent.py:97, self.retry_attempts = retry_attempts). With retry_attempts=0, the while i < self.retry_attempts: loop at critic_agent.py:138 never executes, so control falls through to the warnings.warn(...) + random.choice(...) fallback at lines 165-170.

Fix

Mirror the parent's clamp in the subclass:

self.retry_attempts = max(1, retry_attempts)

This restores the invariant ChatAgent already enforces; clone() also propagates the clamped value since it passes self.retry_attempts on.

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Repository UI

Review profile: CHILL

Plan: Advanced

Run ID: a7548a69-8053-41be-9b2a-e95e3f62f155

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}