RetryOrchestratorSkill

What problem does it solve?

This Skill addresses the challenge of handling errors and failures within an AI agent swarm by providing automated detection and recovery orchestration, preventing task failures from halting progress.

Core Features & Use Cases

• Passive Error Monitoring: Listens for error, failure, timeout, and rejection events on the message bus.
• Intelligent Recovery: Decides on appropriate actions like retrying, refining prompts, swapping skills, reflecting on errors, or escalating to the user.
• Use Case: If a skill fails due to a transient LLM timeout, this Skill can automatically retry the task with exponential backoff, or if a critical security violation occurs, it can escalate the issue for user review.

Quick Start

Instruct the RetryOrchestratorSkill to monitor for task failures and initiate recovery actions.