Description

removeCodeBlocks fails to strip markdown fences from truncated LLM responses. When an LLM returns JSON wrapped in ```json that gets truncated (no closing fence), the regex doesn't match
and raw text with backticks is passed to JSON.parse, causing a SyntaxError. This silently drops all extracted memories.

Fixed by making the closing fence optional in the regex: (?:```|$) matches either the closing fence or end-of-string.

Fixes #4401

Type of change

• Bug fix (non-breaking change which fixes an issue)

How Has This Been Tested?

• Unit Test

Added 7 new test cases covering: truncated blocks (missing closing fence), incomplete JSON, orphan trailing fence, bare fence, multi-block + truncated combo, empty input, and CRLF line endings.
All 12 tests pass (5 existing + 7 new).

cd mem0-ts && npx jest --testPathPattern="remove-code-blocks"

Checklist:

• My code follows the style guidelines of this project
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes
• Any dependent changes have been merged and published in downstream modules
• I have checked my code and corrected any misspellings

Maintainer Checklist

• closes removeCodeBlocks fails to strip markdown fences from truncated LLM responses #4401
• Made sure Checks passed