Crate robit_ai 

robit-ai: LLM API layer for the Robit framework.

Provides a unified LlmClient for interacting with multiple LLM providers through the OpenAI-compatible protocol. Configuration is loaded from .robit/config.toml (project-local) or ~/.robit/config.toml (global).

Re-exports

pub use client::LlmClient;

pub use config::load_config;

pub use config::load_env;

pub use config::AppConfig;

pub use config::BotConfig;

pub use config::ChannelsConfig;

pub use config::ConfirmKeywordsConfig;

pub use config::ContextConfig;

pub use config::ModelConfig;

pub use config::ProviderConfig;

pub use config::QqBotConfig;

pub use config::ResolvedModel;

pub use config::RetryConfig;

pub use config::RobitConfig;

pub use error::LlmError;

Modules

client

LlmClient: a thin wrapper around async-openai with unified config support.

config

Configuration loading for config.toml.

error

Error types for the robit-ai crate.

Structs

ChatCompletionMessageToolCall

ChatCompletionRequestAssistantMessage

ChatCompletionRequestMessageContentPartImage

ChatCompletionRequestMessageContentPartText

ChatCompletionRequestSystemMessage

ChatCompletionRequestUserMessage

CompletionUsage

Usage statistics for the completion request.

CreateChatCompletionResponse

Represents a chat completion response returned by model, based on the provided input.

CreateChatCompletionStreamResponse

Represents a streamed chunk of a chat completion response returned by the model, based on the provided input. Learn more.

Enums

ChatCompletionRequestMessage

ChatCompletionRequestUserMessageContent

ChatCompletionRequestUserMessageContentPart

ChatCompletionTools

Role

Type Aliases

ChatCompletionResponseStream

Parsed server side events stream until an [DONE] is received from server.