Overview
This patch release makes summary generation failures non-blocking by default so transient upstream errors no longer interrupt the active chat. It also adds an explicit operator valve to preserve the old hard-failure behavior when needed.
Bug Fixes
- Graceful summary failure handling (Issue #74): Background summary LLM failures now default to a silent skip instead of re-raising into the active chat flow.
- Chat continuity preserved during transient upstream errors: Short-lived upstream provider failures such as 502s now log the summary error and continue the current chat without saving a summary for that turn.
New Features
SUMMARY_FAIL_MODEvalve: Added a new valve withsilentandraisemodes so operators can choose between chat-friendly degradation and strict failure visibility.- Regression coverage for both modes: Added tests for the default silent path and the opt-in raise path.
Migration Notes
No breaking changes. Default behavior is now SUMMARY_FAIL_MODE="silent". Set SUMMARY_FAIL_MODE="raise" if you need the previous hard-failure behavior for debugging.
Version Changes
Plugin Updates
- Async Context Compression: v1.6.2 → v1.6.3 | 📖 README
Contributors
- @dvystrcil via #76
- @Fu-Jie for maintainer follow-up and release preparation
Related Pull Requests
- fix(async-context-compression): graceful failure on summary LLM error (Shape B, closes #74) (#76) by @dvystrcil