LLM::Adapter
A normalized adapter interface for LLM clients.
Implementations should return a String response (JSON text after any provider-specific cleanup).
Class methods
Instance methods
Send a single prompt and get a response as a String.
A bundle request: like request_with_context, but an adapter that can
tell raises LLM::ContextOverflow instead of returning "", so the
bundle can be re-split rather than lost.
Send chat-style messages (system/user) and get a response as a String.
Request next step using provider-native tool definitions. Implementations that do not support this can fallback to regular JSON-mode requests.
A single system + user request.