Watch the videoPrefer video? Watch Build a multi-agent system with LaunchDarkly Agent Skills for a walkthrough of this tutorial.
What you’ll build
A three-agent pipeline called “Side Project Launcher”:- Idea Validator: researches competitors, analyzes market gaps, scores viability
- Landing Page Writer: generates headlines, copy, and CTAs based on your value prop
- Tech Stack Advisor: recommends frameworks, databases, and hosting based on your requirements
Prerequisites
- LaunchDarkly account (free trial works)
- Claude Code, Cursor, or Windsurf installed
- LaunchDarkly API access token (for creating configs)
- Anthropic API key (for running the sample app)
Start your free trialWant to follow along? Start your 14-day free trial of LaunchDarkly. No credit card required.
30-second quickstart
If you just want to get started, here’s the fastest path: 1. Install skills:Install Agent Skills in Claude Code, Cursor, or Windsurf
Agent Skills work with any editor that supports the Agent Skills specification.Step 1: Install the skills
You have two options: Option A: Use skills.sh (recommended) skills.sh is an open directory for agent skills. Install LaunchDarkly skills with one command:Step 2: Restart your editor
Close and reopen your editor. The skills load on startup. How to verify: Type/configs in Claude Code. You should see autocomplete suggestions. In Cursor, ask “what LaunchDarkly skills do you have?” and the assistant should list them.
Step 3: Set your API token
writer role works, or use a custom role with createProject and createAIConfig permissions.
Build a multi-agent project
Now let’s build something real: a Side Project Launcher that helps you validate ideas, write landing pages, and pick the right tech stack. Tell the assistant:What the assistant creates
The assistant uses several skills automatically:- projects: creates the LaunchDarkly project
- configs-create: builds each agent configuration with variables
- tools: defines tools for function calling

Claude Code creates the configs and provides SDK keys
If the assistant sets a
temperature, top_p, or top_k on a config, remove it. The SDK forwards the config’s model parameters to Anthropic, and the Anthropic API version that current Anthropic SDKs use rejects these parameters, which fails the run with a 400 error.{{idea}}, {{target_audience}}, etc.) get filled in at runtime when you call the SDK. That’s how each user gets personalized output.
What it looks like in LaunchDarkly

AgentControl configs list showing the three agents created by Agent Skills
- 3 configs with instructions, model settings, and variables
- 3 tools with parameter definitions ready for function calling
- Default targeting serving the configuration to all users

Default targeting serves the configuration to all users

Idea validator config with instructions, variables, and tools
Run the Side Project Launcher
The full working code is available on GitHub: launchdarkly-labs/side-project-researcher Clone it and run:launchdarkly-ai-server>=1.0.0 and
launchdarkly-ai-langchain-agents>=1.0.0. The handler packages version independently of the core SDK.
Pin each one on its own rather than expecting one version number to cover both.
You’ll need both the LaunchDarkly SDK key (from your project’s SDK settings) and your Anthropic API key in the .env file. The assistant can surface the SDK key from your project details, but store it in .env rather than hardcoding it.
The app prompts you for your idea details:

The app prompts you for your side project details

Idea validator output with market analysis

Tech stack advisor recommending frameworks and infrastructure
Connect to your framework
The config stores your model, instructions, and tools. The SDK fetches the config and handles variable substitution automatically.Code snippets show the patternThe snippets below show the integration pattern. They omit error handling and the tool implementations for brevity. For complete, runnable code, use the sample repo.
Initialize the SDK
init_client() is async. Call it once inside your entry point, before the graph runs. You’ll see it in main() at the end of the next section.
Build a context and check a config is being served
Wire it to LangGraph
LangGraph orchestrates multi-agent workflows as a graph of nodes, but you can use any orchestrator: CrewAI, LlamaIndex, Bedrock AgentCore, or custom code. To compare options, read Compare AI orchestrators. By wiring configs to each node, your agents fetch their model, instructions, and tools dynamically from LaunchDarkly. This lets you swap models within a provider (for example, Sonnet to Haiku), update prompts, or disable agents without redeploying.Tools require runtime handlersThe config defines tool schemas, but your code must implement the actual tool functions. Pass them to
config() as tool_handlers, a dict keyed by tool name. A config tool with no matching handler isn’t offered to the model, which means the agent can’t call it. Implement a handler for every tool you attach. The sample repo’s tools.py defines TOOL_HANDLERS for the tools these configs attach.What you can do next
After your agents are in LaunchDarkly:- A/B test variations: Split traffic between prompt variations or model sizes (for example, Sonnet versus Haiku) to see which performs better.
- Target by segment: Premium users get one variation, free users get another.
- Kill switch: Disable a misbehaving agent instantly from the UI.
- Track costs: Monitor tokens and latency per variation.
Troubleshooting
Skills installed but not working: Restart your editor after installing skills. They load on startup. “Permission denied” errors: Check that your API token hascreateProject and createAIConfig permissions. The writer role includes both.
Config comes back disabled: Your targeting rules may not match the context you’re passing. Check that default targeting is enabled, or that your context attributes match your rules.
Tools defined but not executing: The config defines tool schemas, but your code must implement handlers. See the sample repo for tool binding examples.
Can’t find SDK key: After Agent Skills creates your project, find the SDK key in your project’s Settings > Environments > SDK key. Copy it to your .env file.
FAQ
Do I need Claude Code, or does this work in Cursor/Windsurf?
Agent Skills work in any editor that supports the Agent Skills specification. This includes Claude Code, Cursor, and Windsurf. The installation process is the same.What’s the difference between Agent Skills and the MCP server?
Both give your AI assistant access to LaunchDarkly. Agent Skills are text-based playbooks that teach the assistant workflows. The MCP server exposes the LaunchDarkly API as tools. You can use either or both.What permissions does my API token need?
Thewriter role works, or use a custom role with createProject and createAIConfig permissions.
Where do I see the created configs?
In the LaunchDarkly UI: go to your project, then click Agent management > Agents in the left sidebar. Each config shows its instructions, model, tools, and targeting rules.How do I delete or reset generated configs?
In the LaunchDarkly UI, open the config and delete it, as described in Delete configs. Deleting a config is permanent. Or ask the assistant: “Delete the config called idea-validator in project side-project-launcher.”Can I use this with frameworks other than LangGraph?
Yes. The SDK returns model name, instructions, and tools as data. You wire that into whatever framework you use: CrewAI, LlamaIndex, Bedrock AgentCore, or custom code.Does this work for completion mode (chat) or just agent mode
Both, and the calling code is the same either way.config(key=..., handler=...).invoke(...) covers both modes. What changes is the handler you give it. A messages handler such as create_langchain_messages_handler() serves completion mode, and an agents handler such as create_langchain_agents_handler() serves agent mode. A handler declares which modes it can serve, which means pairing one with the wrong config raises rather than running the wrong shape. To learn more, read Agent mode vs completion mode.
Next steps
- Read the Python AI SDK Reference for detailed SDK usage
- Try building a data extraction pipeline to deploy configs with Vercel
- Beyond n8n for Workflow Automation: Agent Graphs - Externalize the multi-agent topology you just built into a visual graph
- Building Framework-Agnostic AI Swarms - Run the same generated configs across LangGraph, Strands, and OpenAI Swarm
- Offline Evaluation of RAG-Grounded Answers - Add a regression-testing pass for the agents you generated
- When to add online evals - Decide when to attach judges to your generated configs in production