AE-T35FP6NY
Summarization middleware deletes agent conversation history and stores the provider error text as the summary when the summary call fails
MEMORY FAILUREseverity: CRITICALcause: LIKELYoutcome: RESOLVED UNVERIFIEDconfidence: LOW
When the summarization model call raised (e.g. a 429), SummarizationMiddleware caught the exception, returned the error string as the summary and still emitted a remove-all-messages update, so most of the thread history was permanently deleted and replaced by the error text. With a checkpointer the corrupted state is persisted. The merged fix retries the summary call and lets the error propagate instead of fabricating a summary.
- Framework / agent
- LangChain · create_agent tool-calling agent with summarization middleware
- Remediation attempts
- TESTEDSUGGESTED
- Recurrence
- not documented
- Source languages
- en
- Updated
- 2026-09-30
Sources
- GITHUB ISSUE SummarizationMiddleware deletes conversation history when the summary model call fails, inserting the error text as the summary — github.com/langchain-ai/langchain, retrieved 2026-09-29
- GITHUB PULL REQUEST fix(langchain): preserve history on `SummarizationMiddleware` summary failure — github.com/langchain-ai/langchain, retrieved 2026-09-29
- GITHUB PULL REQUEST fix(langchain): preserve history when summary generation fails — github.com/langchain-ai/langchain, retrieved 2026-09-29
Symptoms
- 8 of 10 messages are removed from the thread when the summary model call fails
- The inserted summary message contains the provider error text instead of a summary, and no exception is raised
The full record — root-cause evidence, every remediation attempt with its status and verification, failed attempts, patch references, verbatim quotes and recurrence — is a paid lookup (0.018 USDC via x402). Agents:
GET /api/v1/cases/AE-T35FP6NY. PricingSimilarity to your system is not implied. A remediation that worked in the documented context may not work in yours.