Structs§
- Anthropic
Prompt Caching Middleware - Anthropic-specific prompt caching middleware. Marks system prompts for caching to reduce latency on subsequent requests with the same base prompt.
- Base
System Prompt Middleware - Deep
Agent Prompt Middleware - Deep Agent prompt middleware that injects comprehensive tool usage instructions and examples to force the LLM to actually call tools instead of just talking about them.
- Filesystem
Middleware - Hitl
Policy - Human
InLoop Middleware - Middleware
Context - Read/write state handle exposed to middleware implementations.
- Model
Request - Request sent to the underlying language model. Middlewares can augment the system prompt or mutate the pending message list before the model call.
- Planning
Middleware - SubAgent
Descriptor - SubAgent
Middleware - SubAgent
Registration - Summarization
Middleware - Task
Router Tool
Traits§
- Agent
Middleware - Middleware hook that can register additional tools and mutate the model request prior to execution. Mirrors the Python AgentMiddleware contracts but keeps the interface async-first for future network calls.