Cap OpenAI SDK output-token limits (OWASP LLM10)

Proprietary

Reduce explicit limits above the selected maximum in OpenAI chat-completion request builders, including legacy maxTokens and maxCompletionTokens. Smaller, dynamic, missing, and nonpositive limits are preserved. Lower budgets can truncate output or conflict with provider minimums.

org.openrewrite.ai.security.LimitOpenAiSdkOutputTokens

Required

maximumTokensint

A positive, application-selected ceiling for explicit output-token limits. Larger literal limits are reduced to this value. Smaller limits are preserved. Choose a budget appropriate for the model and workload: reducing it can truncate responses, reasoning, structured output, or tool calls.

mod run . --recipe=org.openrewrite.ai.security.LimitOpenAiSdkOutputTokens \
  -P maximumTokens=4096

Install it once from Recipe actions, then run it on every repository under the current directory. New to the Moderne CLI? Set it up.