Jido.AI.Reasoning.ReAct.Config
(Jido AI v2.1.0)
View Source
Canonical configuration for the Task-based ReAct runtime.
Summary
Functions
Returns a stable config fingerprint used by checkpoint tokens.
Convert config to generation options for ReqLLM.Generation.stream_text/3
and ReqLLM.Generation.generate_text/3.
Merge request-scoped LLM option overrides into an existing normalized option list.
Build a runtime config from options.
Convert runtime tools to ReqLLM tool definitions.
Returns the effective stream consumer timeout.
Types
@type t() :: %Jido.AI.Reasoning.ReAct.Config{ effect_policy: any(), llm: %{ :max_tokens => integer(), :temperature => number(), :req_http_options => [any()], :llm_opts => any(), optional(:timeout_ms) => nil | integer(), tool_choice: any() }, max_iterations: integer(), model: any(), observability: %{ emit_telemetry?: boolean(), emit_signals?: boolean(), redact_tool_args?: boolean() }, pending_input_server: nil | nil | any(), request_transformer: nil | nil | atom(), stream_timeout_ms: integer(), streaming: boolean(), system_prompt: nil | nil | binary(), token: %{ :secret => binary(), optional(:ttl_ms) => nil | integer(), compress?: boolean() }, tool_exec: %{ max_retries: integer(), timeout_ms: integer(), retry_backoff_ms: integer(), concurrency: integer() }, tools: map(), trace: %{ capture_deltas?: boolean(), capture_thinking?: boolean(), capture_messages?: boolean() }, version: integer() }
Functions
Returns a stable config fingerprint used by checkpoint tokens.
Convert config to generation options for ReqLLM.Generation.stream_text/3
and ReqLLM.Generation.generate_text/3.
Merge request-scoped LLM option overrides into an existing normalized option list.
Build a runtime config from options.
@spec reqllm_tools(t()) :: [ReqLLM.Tool.t()]
Convert runtime tools to ReqLLM tool definitions.
@spec stream_timeout(t()) :: pos_integer()
Returns the effective stream consumer timeout.
When stream_timeout_ms is 0 (default), auto-derives from
tool_exec.timeout_ms + 60_000 to ensure the stream consumer
outlives the longest possible tool execution plus LLM response time.