Team
Teams coordinate multiple agents working together using different execution strategies.
Specification
apiVersion: ark.mckinsey.com/v1alpha1
kind: Team
metadata:
name: example-team
spec:
# Team members - agents or nested teams
members:
- name: researcher
type: agent
- name: analyst
type: agent
- name: writer
type: agent
# Execution strategy - how members collaborate
strategy: selector # Options: sequential, selector, graph
# Enable looping for sequential strategy (optional)
# loops: true
# maxTurns: 10 # Required when loops is true
# Selector configuration - for strategy: selector
selector:
agent: planner # Agent to use for selection (required)
selectorPrompt: "Choose the best agent for: {{.Input}}" # Optional
# Graph constraints (optional) - can be combined with selector strategy
# When combined with selector, limits AI selection to valid graph transitions
graph:
edges:
- from: researcher
to: analyst
- from: analyst
to: writer
# # Sequential configuration - for strategy: sequential
# strategy: sequential
# # Single pass through members, no maxTurns needed
# # Sequential with loops - for cycling through members
# strategy: sequential
# loops: true
# maxTurns: 10 # Required - prevents infinite loops
# # Graph-only configuration - for strategy: graph
# strategy: graph
# maxTurns: 10
# graph:
# edges:
# - from: researcher
# to: analyst
# - from: analyst
# to: writerExecution Strategies
Execution strategies define how team members are chosen during query execution.
- sequential - Agents process input one after another in a single pass
- sequential + loops - Agents cycle through members repeatedly until
maxTurnsor terminate - selector - Dynamic agent selection based on criteria, LLM chooses the next agent for the job
- graph - Custom execution flows with edges, supports more complex workflows
- selector + graph - Combines AI-driven selection with workflow constraints (selector agent chooses from graph-defined valid transitions)
strategy: round-robin is deprecated and will be removed in v1.0.0. Use strategy: sequential with loops: true instead. Existing round-robin teams are automatically migrated on create/update.
Turn Limiting
The maxTurns field limits execution turns. Its usage depends on the strategy:
- sequential - Not applicable for single pass. Required when
loops: trueto prevent infinite loops. - selector - Limits selection rounds (each round = one agent selection and execution)
- graph - Limits edge traversals through the execution graph
When maxTurns is reached:
- Team execution stops gracefully
- All responses generated up to the limit are returned
- Warning event emitted:
TeamMaxTurnsReached - Query completes successfully (not an error)
Last updated on