> ## Documentation Index
> Fetch the complete documentation index at: https://praison.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# runtime • AI Agent SDK

> Runtime execution components and capability system for PraisonAI agents.

# runtime

AI Agent

Runtime execution components and capability system for PraisonAI agents.

This module provides runtime execution abstractions, protocols, and capability
validation for standardizing agent execution across different harness types.

Key Components:

* Runtime execution contexts: PreparedTurnContext, TurnRuntimeProtocol
* Agent runtime registry: AgentRuntimeProtocol, RuntimeRegistry
* Tool result middleware: RuntimeToolResultMiddleware, NormalizedToolResult
* Registry patterns for runtime resolution and middleware management
* Doctor migration protocol: DoctorContractProtocol for config migration
* Capability validation: RuntimeCapability, RuntimeCapabilityMatrix, validate\_capabilities
* Runtime protocols: AgentRuntimeProtocol for capability reporting
* Turn-time runtime resolution for handoffs and sub-agents

Features:

* Turn-based runtime execution with PreparedTurnContext
* RuntimeCapability flags for feature detection
* RuntimeCapabilityMatrix for runtime capability sets
* Capability validation at runtime selection time
* Support for native, plugin harness, and managed runtimes
* Dynamic runtime resolution at handoff boundaries

## Import

```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
from praisonaiagents import runtime
```

### Constants

| Name | Value |
| - | - |
| `_LAZY_GROUPS` | `{'turn_context': {'PreparedTurnContext': ('praisonaiagents.runtime.turn_context', 'PreparedTurnContext'), 'ModelReference': ('praisonaiagents.runtime.turn_context', 'ModelReference'), 'ToolSchema': ('...` |


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.