llm-thinking-strip-ollama-only

IN premise

Thinking-marker stripping (`Thinking...` / `...done thinking.`) is applied only to Ollama model output; Claude and Gemini output passes through unmodified even if it contains the same markers.

Summary

The code that removes "Thinking..." and "...done thinking." markers from LLM responses only runs on output from Ollama models. If Claude or Gemini happened to produce those same markers in their output, they would be left in place and could end up visible to users or downstream consumers.

Details

Source

entries/2026/05/05/tests-test_llm.md