Collected sources and patterns will appear here. Add from search or the patterns library.
LLMResponse -> SanitizedLLMResponse
Optionally suppress auto-extraction and auto-running of fenced code blocks from LLM responses when explicit execution tools are present.
Problem it solves
Preventing duplicate or unauthorized execution of code embedded in LLM output when formal execution tools are also accessible to the agent.
Consumes
Emits
The real projects this mechanism was found in. Attribution is the point — this is how the best teams actually do it.