Skip to main content

session_affinity_key_enabled

vtcode_config::core::prompt_cache

Function session_affinity_key_enabled 

Source 
pub fn session_affinity_key_enabled(
    provider_name: &str,
    global_prompt_cache_enabled: bool,
    openai_prompt_cache_enabled: bool,
) -> bool
Expand description

Whether LLMRequest.prompt_cache_key should carry session lineage for this provider.

OpenAI/Merge keep the OpenAI prompt-cache enablement gate. OpenRouter/xAI/OpenCode Go need lineage for documented session affinity even when the OpenAI-specific cache block is off (global prompt-cache master switch still applies). OpenCode Go rejects requests without its session header, so its lineage must not depend on the OpenAI cache flag.