Concepts
Workflows Taxonomy
Understanding the six categories of workflow patterns
Overview
Workflows are operational execution patterns that define how tasks are executed during the Execution/Evaluation phases of the Agent Lifecycle. They are macro-level orchestration patterns, distinct from the Agent Loop (micro-execution) and the Lifecycle (system layer).
The Six Categories
1. System/Control Workflows
Govern agent management at the platform level. The primary workflow is the Agent Lifecycle: Registration → Discovery → Execution → Evaluation.
2. Task Workflows
Operational patterns for executing work:
- Routing: Classify inputs and direct to specialized tasks
- Prompt Chaining: Sequential steps with validation gates
- Orchestrator-Workers: Central orchestrator delegates to workers
- Parallelization: Simultaneous execution with aggregation
- Evaluator-Optimizer: Generate-evaluate-refine loops
Reference: Anthropic: Building Effective Agents
3. Quality Workflows
Ensure outputs meet standards:
- Rules Validation: Defined criteria and constraints
- Visual Checks: Screenshots and renders
- LLM-as-Judge: Model-based evaluation
4. Recovery Workflows
Handle failures and errors:
- Retries: Automatic retry mechanisms
- Fallback: Alternative paths when primary fails
- Timeouts: Long-running operation handling
5. Human-in-the-Loop Workflows
Integrate human oversight:
- Approval Workflows: Human approval before proceeding
- Manual Delegation: Human task assignment
6. Multi-Agent Workflows
Coordinate multiple agents:
- Agent Coordination: Multiple agents working together
- Distributed Execution: Tasks distributed across agents
Key Distinctions
- Workflows are macro-level orchestration patterns used during Execution/Evaluation
- Agent Loop is the micro-level cognitive cycle inside workflows
- Lifecycle is the system-level governance around workflows
Next Steps
- Explore specific Task Workflows
- Understand Quality Assurance mechanisms
- Learn about Recovery Patterns
- Read the full specification in AGENTS.md Section 3