Copilot Chat in VS Code: orchestration loop/stall with tool schema mismatch and virtual-tools drops

[alpet83]

🏷️ Discussion Type

Bug

💬 Feature/Topic Area

VS Code

Body

Copilot Chat Agent Loop in VS Code (No UI Freeze): Repeated Patch/Analysis Cycle + Tool Orchestration Errors

Summary

I am reporting a Copilot Chat issue in VS Code where the UI remains responsive, but Agent Chat enters a repeat loop:

• re-applies edits to the same files
• repeats analysis of the same state
• does not converge to completion

This is not the typical full VS Code freeze scenario.

Environment

• OS: Windows
• VS Code: [your version]
• GitHub Copilot Chat extension: 0.40.1
• Workspace: multi-root
• Mode: Agent chat with tool usage

What I Observe

1. VS Code window stays responsive.
2. Agent repeatedly cycles over the same two script edits and follow-up analysis.
3. Loop ends only after manual interruption/retry.

Why I Think This Is Not Just Rate Limit

From logs, the dominant pattern is orchestration/tooling-related:

• repeated 400 tool schema/tag mismatch:
  • Input tag 'tool_search_tool_regex_20251119' ... does not match expected tags
• frequent:
  • [ToolCallingLoop] Stop hook result: shouldContinue=false
• repeated:
  • [virtual-tools] Had to drop 1 tools due to limit constraints
• occasional server 500 + retry

I do see rate-limit-like behavior at times, but it looks secondary to the tool orchestration loop.

Minimal Repro Pattern

1. Open Agent Chat in VS Code.
2. Run a multi-step editing workflow involving iterative patching and analysis.
3. After several rounds, Agent begins repeating edits/analysis over the same files.

Impact

• Long-running coding workflows become unreliable.
• Significant productivity loss due to manual loop-break/restart.
• Hard to trust completion state in autonomous sessions.

Mitigation Tried

• Added strict orchestration timeouts on our side (per-step and overall), which prevents indefinite hangs.
• This mitigates symptoms but does not fix the upstream loop behavior.

Attachments

• Sanitized log bundle attached (.zip) including:
  • GitHub Copilot Chat.log excerpts
  • agentSessionsOutput.log excerpts
  • incident summary file

Request

Please investigate Copilot Chat agent orchestration for:

1. tool schema/tag mismatch handling
2. virtual-tools constraint/drop behavior
3. loop prevention / convergence safeguards for repeated patch-analysis cycles

---

If helpful, I can provide:

• exact timestamps (UTC/local)
• request IDs from failed turns
• a shorter deterministic repro sequence.
  Best regards,
  Aleksandr Petrov
  copilot-log-report-20260325-171343.zip

[github-actions[bot]]

💬 Your Product Feedback Has Been Submitted 🎉

Thank you for taking the time to share your insights with us! Your feedback is invaluable as we build a better GitHub experience for all our users.

Here's what you can expect moving forward ⏩

• Your input will be carefully reviewed and cataloged by members of our product teams. 
  • Due to the high volume of submissions, we may not always be able to provide individual responses.
  • Rest assured, your feedback will help chart our course for product improvements.
• Other users may engage with your post, sharing their own perspectives or experiences. 
• GitHub staff may reach out for further clarification or insight. 
  • We may 'Answer' your discussion if there is a current solution, workaround, or roadmap/changelog post related to the feedback.

Where to look to see what's shipping 👀

• Read the Changelog for real-time updates on the latest GitHub features, enhancements, and calls for feedback.
• Explore our Product Roadmap, which details upcoming major releases and initiatives.

What you can do in the meantime 💻

• Upvote and comment on other user feedback Discussions that resonate with you.
• Add more information at any point! Useful details include: use cases, relevant labels, desired outcomes, and any accompanying screenshots.

As a member of the GitHub community, your participation is essential. While we can't promise that every suggestion will be implemented, we want to emphasize that your feedback is instrumental in guiding our decisions and priorities.

Thank you once again for your contribution to making GitHub even better! We're grateful for your ongoing support and collaboration in shaping the future of our platform. ⭐