Agent-Skills.md

Agent Skills: LangGraph Routing Skill

Conditional edge routing and state-based transitions for LangGraph workflows

UncategorizedID: a5c-ai/babysitter/langgraph-routing

Repository

a5c-ai/babysitter
a5c-aiLicense: MIT
244

Install this agent skill to your local

pnpm dlx add-skill https://github.com/a5c-ai/babysitter/tree/HEAD/library/specializations/ai-agents-conversational/skills/langgraph-routing

Skill Files

Browse the full folder contents for langgraph-routing.

Download Skill

Loading file tree…

library/specializations/ai-agents-conversational/skills/langgraph-routing/SKILL.md

Skill Metadata

Name
langgraph-routing
Description
Conditional edge routing and state-based transitions for LangGraph workflows

LangGraph Routing Skill

Capabilities

  • Design conditional edge routing in LangGraph
  • Implement state-based transition logic
  • Create dynamic routing functions
  • Handle multi-path workflow branches
  • Implement router nodes for complex decisions
  • Design fallback and error routing paths

Target Processes

  • langgraph-workflow-design
  • plan-and-execute-agent

Implementation Details

Routing Patterns

  1. Conditional Edges: add_conditional_edges with routing functions
  2. Router Nodes: Dedicated nodes for routing decisions
  3. State-Based Routing: Routing based on state values
  4. LLM-Based Routing: Using LLM to determine next node

Configuration Options

  • Routing function definitions
  • Path mapping configurations
  • Default/fallback routes
  • Cycle detection settings
  • Max iteration limits

Best Practices

  • Clear routing logic documentation
  • Handle all possible states
  • Implement fallback paths
  • Avoid infinite cycles
  • Use descriptive edge names

Dependencies

  • langgraph