Crate brainos_cortex 

Brain Cortex

Reasoning core providing:

• LLM client (Ollama, OpenAI compatible)
• Hybrid provider with trait-based adapter pattern
• Context assembly from memory + user model
• Token budget management
• Tool calling and action dispatch
• Structured output validation with retry logic

Re-exports

pub use llm::create_provider;

pub use llm::LlmError;

pub use llm::LlmProvider;

pub use llm::Message;

pub use llm::OllamaProvider;

pub use llm::OpenAiProvider;

pub use llm::ProviderConfig;

pub use llm::Response;

pub use llm::ResponseChunk;

pub use llm::Role;

pub use llm::Usage;

Modules

actions

Action dispatch — tool execution.

context

Context assembly — builds LLM prompts with token budget management.

llm

LLM client — hybrid provider with trait-based adapter.