Help us improve
Share bugs, ideas, or general feedback.
From launchdarkly
Creates and manages LaunchDarkly agent graphs — directed graphs of configs with handoff logic for multi-agent workflows. Use when building routing between configs.
npx claudepluginhub launchdarkly/ai-toolingHow this skill is triggered — by the user, by Claude, or both
Slash command
/launchdarkly:agent-graphsThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
You're using a skill that will guide you through creating and managing agent graphs in LaunchDarkly. Your job is to design the graph topology, create it with the right edges and handoffs, and verify the routing between config nodes.
Builds production-grade stateful multi-actor AI agents with LangGraph, covering graph construction, state management, persistence, cycles, branches, human-in-the-loop, and ReAct patterns.
Build agentic workflow pipelines in Python combining deterministic flow control with AI reasoning via Studio Agent and MCP tools.
Guides architectural decisions for LangGraph applications. Use when choosing between LangGraph vs alternatives, designing state schemas with reducers, structuring graphs with subgraphs, or selecting persistence and streaming approaches.
Share bugs, ideas, or general feedback.
You're using a skill that will guide you through creating and managing agent graphs in LaunchDarkly. Your job is to design the graph topology, create it with the right edges and handoffs, and verify the routing between config nodes.
This skill requires the remotely hosted LaunchDarkly MCP server to be configured in your environment.
Required MCP tools:
create-agent-graph -- create a new graph with nodes and edgesget-agent-graph -- inspect a graph's structure and edgeslist-agent-graphs -- browse existing graphs in the projectOptional MCP tools:
update-agent-graph -- modify edges, root config, or descriptiondelete-agent-graph -- permanently remove a graphget-ai-config -- inspect individual configs that serve as nodescreate-ai-config -- create new configs to use as graph nodesAn agent graph is a directed graph where:
| Scenario | Example |
|---|---|
| Multi-step workflows | Triage agent -> Specialist agent -> Summary agent |
| Routing by intent | Router agent decides which specialist handles the request |
| Escalation chains | L1 support -> L2 support -> Human handoff |
| Pipeline processing | Extract -> Transform -> Validate -> Store |
[Root Config] --edge--> [Config A] --edge--> [Config C]
\--edge--> [Config B]
Each edge has:
key -- unique identifier for the edgesourceConfig -- the config key that routes FROMtargetConfig -- the config key that routes TOhandoff (optional) -- data/instructions passed during the transitionBefore creating anything:
list-agent-graphs to avoid duplicatesget-ai-config to see what nodes already existEach node in the graph must be an existing config. If configs don't exist yet:
create-ai-config to create each agent configget-ai-configUse create-agent-graph with:
projectKey -- the project containing the configskey -- unique identifier for the graphname -- human-readable display namedescription (optional) -- explain the graph's purposerootConfigKey -- the entry-point config keyedges -- array of connections between configs{
"projectKey": "my-project",
"key": "support-triage-graph",
"name": "Customer Support Triage",
"description": "Routes customer queries to the appropriate specialist agent",
"rootConfigKey": "triage-agent",
"edges": [
{
"key": "triage-to-billing",
"sourceConfig": "triage-agent",
"targetConfig": "billing-specialist",
"handoff": {"category": "billing", "priority": "normal"}
},
{
"key": "triage-to-technical",
"sourceConfig": "triage-agent",
"targetConfig": "technical-specialist",
"handoff": {"category": "technical", "priority": "normal"}
}
]
}
get-agent-graph to confirm the graph was created with the correct structureReport results:
| Situation | Action |
|---|---|
| Config doesn't exist yet | Create it first with create-ai-config before referencing in a graph |
| Circular routing | Allowed but warn user — ensure there's a termination condition in the agent logic |
| Single-node graph | Valid but unusual — consider if a graph is actually needed |
| Updating edges | Use update-agent-graph — provide the complete new edge list |