Skip to content

[`agentic_crew.runners.strands_runner`](#module-agentic_crew.runners.strands_runner)

AWS Strands runner implementation.

Strands is lightweight and AWS-native, good for:

  • Simple agent workflows
  • AWS Bedrock integration
  • Minimal dependencies
  • Plain Python function tools
StrandsRunnerRunner that uses AWS Strands for crew execution.

class agentic_crew.runners.strands_runner.StrandsRunner

Section titled “class agentic_crew.runners.strands_runner.StrandsRunner”

Bases: agentic_crew.runners.base.BaseRunner

Runner that uses AWS Strands for crew execution.

Initialize Strands runner.

‘strands’

Build a Strands agent from configuration.

Strands uses a single-agent model, so we combine crew tasks into a system prompt for one capable agent.

Args: crew_config: Universal crew configuration.

Returns: Strands Agent object.

Get Strands-compatible model provider from LLM config.

Args: llm_config: LLM configuration from crew config.

Returns: Model provider string for Strands, or None for default.

Execute the Strands agent.

Args: crew: Strands Agent object. inputs: Inputs for the agent.

Returns: Agent output as string.

Build a Strands agent.

Args: agent_config: Agent configuration. tools: Optional tools.

Returns: Strands Agent object.

Build a task for Strands.

In Strands, tasks are prompts to the agent. Returns a dict that can be used to construct prompts.

Args: task_config: Task configuration. agent: Agent for the task.

Returns: Task configuration dict.

Build a comprehensive system prompt from crew config.

Combines all agent roles, goals, and backstories into a single capable system prompt.

Args: crew_config: Crew configuration.

Returns: System prompt string.

Collect tools from crew configuration.

Args: crew_config: Crew configuration.

Returns: List of tool functions.