litellm_transport.py DOX

Purpose

Own Agent Zero's LiteLLM transport adapter for Chat Completions and Responses API calls.
Normalize Agent Zero model-call kwargs into provider-safe LiteLLM requests.
Preserve canonical response metadata for history, provider-state continuation, and fallback decisions.

litellm_transport.py owns the runtime implementation.
litellm_transport.py.dox.md owns durable notes about responsibilities, contracts, side effects, and verification for that implementation.
Classes:
TransportMode
TransportRecovery
TransportPolicy
LiteLLMTransport
ChatCompletionsTransport
ResponsesTransport
ResponsesEventParser
Top-level functions include transport cache reset, request normalization, parsing, prompt-cache preparation, and response/error classifiers.

Keep provider selection and provider-specific defaults outside this helper; callers pass a resolved LiteLLM model name and kwargs.
Strip Agent Zero internal kwargs before sending requests to LiteLLM.
Do not send orphan tool controls when no tools are present; strict OpenAI-compatible servers can reject empty tools arrays.
Prefer Responses API when configured, but fallback to Chat Completions when the provider does not support Responses.
Preserve provider-state metadata when Responses API calls succeed, and fall back to local replay when provider state is unsupported.
Keep prompt-cache markers only for providers that accept them.

Add provider-agnostic request cleanup here when multiple OpenAI-compatible providers can benefit.
Treat fallback behavior as a shared transport contract, not a provider registry.
Keep tool conversion symmetric between Chat Completions and Responses requests.

Run pytest tests/test_stream_tool_early_stop.py tests/test_responses_architecture.py -q after changing transport normalization or fallback behavior.
Run local-provider smoke checks when changing OpenAI-compatible request cleanup.

No child DOX files.