feat(auto-title-sync): auto-memory linked chat title refresh + observability

[Ferrum360]

What Problem This Solves

Currently, chat titles in QwenPaw are static placeholders (truncated first message) and never update as the conversation evolves. When auto-memory generates new memory entries, the session title does not reflect the updated topic, making it harder to scan chat history and understand context at a glance.

This PR adds an auto-title-sync mechanism that regenerates the chat title after auto-memory flushes, ensuring the title stays aligned with the actual conversation topic.

Evidence

• Added title_generator.py with refresh_title_after_auto_memory() that generates a concise title from recent messages after auto-memory succeeds.
• Updated middlewares.py so _flush_auto_memory calls title_refresh_callback on success, instead of silently completing.
• Updated manager.py to persist the refreshed title and record auto_title_refresh metadata in chat.meta.
• Added observability in _coordinator.py (INFO log) so title refresh events are visible in logs.
• Adapted memory managers to accept and invoke title_refresh_callback.

Summary of changes

• feat(auto-title-sync): auto-memory linked chat title refresh + observability
• Fix middlewares.py: call title_refresh_callback after _flush_auto_memory success
• Integrate title refresh into manager.py for chat.meta persistence
• Add observability support in _coordinator.py (INFO logging)
• Adapt memory managers to title_refresh_callback interface

Test plan

• Verify chat title auto-refreshes after auto-memory triggers
• Verify title_refresh_callback is only called after successful auto-memory flush
• Verify chat.meta["auto_title_refresh"] is recorded correctly

[github-actions]

Welcome to QwenPaw! 🐾

Hi @Ferrum360, this is your 6th Pull Request.

📋 About PR Template

To help maintainers review your PR faster, please make sure to include:

• ✅ Description - What this PR does and why
• ✅ Type of Change - Bug fix / Feature / Breaking change / Documentation / Refactoring
• ✅ Component(s) Affected - Core / Console / Channels / Skills / CLI / Documentation / Tests / CI/CD / Scripts
• ✅ Checklist:
  • Run and pass pre-commit run --all-files
  • Run and pass relevant tests (pytest or as applicable)
  • Update documentation if needed
• ✅ Testing - How to test these changes
• ✅ Local Verification Evidence:

  pre-commit run --all-files
  # paste summary result
  
  pytest
  # paste summary result

Complete PR information helps speed up the review process. You can edit the PR description to add these details.

🙌 Join Developer Community

Thanks so much for your contribution! We'd love to invite you to join the official QwenPaw developer group! You can find the Discord and DingTalk group links under the "Developer Community" section on our docs page:
https://qwenpaw.agentscope.io/docs/community

We truly appreciate your enthusiasm—and look forward to your future contributions! 😊

We'll review your PR soon.

---

Tip

⭐ If you find QwenPaw useful, please give us a Star!

Staying ahead

Star QwenPaw on GitHub and be instantly notified of new releases.

Your star helps more developers discover this project! 🐾

[jinliyl]

Thank you for working on automatic chat title refresh. I have one architectural concern about the current implementation.

Chat title generation and persistence are application/chat concerns, not memory concerns. Passing title_refresh_callback through BaseMemoryManager and then into MemoryMiddleware couples the memory abstraction to chat presentation and persistence behavior. It also requires every memory backend, including ADBPGMemoryManager and ReMeLightMemoryManager, to accept and forward a parameter that has nothing to do with memory storage, retrieval, summarization, or lifecycle management.

Could we please avoid changing BaseMemoryManager and MemoryMiddleware for this feature?

I suggest introducing a dedicated middleware, for example ChatTitleRefreshMiddleware, backed by a ChatTitleRefreshService. The new middleware could observe the appropriate conversation lifecycle hook, apply its own refresh policy, and delegate title generation and compare-and-set persistence to the chat service. It should then be registered from the application/runtime assembly layer, where both chat services and middleware composition are available.

A possible separation would be:

• MemoryMiddleware: memory search, auto-memory collection, and memory flush behavior only.
• ChatTitleRefreshMiddleware: decide when a completed conversation turn should trigger a title refresh.
• ChatTitleRefreshService: generate the title, locate the chat, preserve manual renames, and update ChatSpec.name and metadata.
• Application/runtime builder: construct and register the middleware with its required chat-layer dependencies.

This would remove the cross-layer callback chain from Workspace to concrete memory managers to BaseMemoryManager to MemoryMiddleware, keep memory backends independent of chat UI behavior, and allow title refresh cadence and configuration to evolve independently from auto-memory.

If the product requirement is specifically to refresh only after a successful auto-memory flush, a generic application-level AutoMemoryFlushed event could be published and handled by the chat title service. Even in that case, I would prefer not to expose a title-specific callback on the memory manager abstraction.

Would you please consider restructuring this part before merging? Thank you.