aiconfig-create

Name: aiconfig-create
Availability: InStock
Author: LaunchDarkly

Create LaunchDarkly AI Configs via a two-step API workflow.

25|8|Updated Feb 3, 2026

Search Tags:#api #verification #tools #launchdarkly #ai-configs #variations #config-creation

Ready to run

npx skills add https://github.com/launchdarkly/agent-skills --skill aiconfig-create-launchdarkly

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: aiconfig-create
Source: https://github.com/launchdarkly/agent-skills/tree/main/skills/ai-configs/aiconfig-create
Command: npx skills add https://github.com/launchdarkly/agent-skills --skill aiconfig-create-launchdarkly

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This skill guides you through setting up AI Configs in LaunchDarkly, helping you choose agent vs completion mode, create the config and its variations, and verify the result.

Core Features & Use Cases

Understand the Use Case First: identify the framework and tooling, decide if agent or completion mode is appropriate.
Two-Step Creation: create the AI config first, then add variations (model, prompts, parameters) and later attach tools.
Verification: fetch the config via API to confirm correct creation and UI visibility.

Quick Start

Copy this skill into your agent client's skills path to begin creating an AI Config and its variations.

Frequently Asked Questions about aiconfig-create

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I create LaunchDarkly AI Configs using the API?▼

Creating LaunchDarkly AI Configs requires a two-step API workflow: first create the base config with a modelConfigKey, then add variations containing the model, prompts, and parameters.

What is the difference between agent and completion mode for AI Configs?▼

Agent and completion modes define the framework behavior for your AI Config. You select between them by identifying your specific use case, tooling, and whether agent or completion mode best fits your project requirements.

Do I need a modelConfigKey when setting up AI Configs in LaunchDarkly?▼

Yes, a modelConfigKey is strictly required when setting up LaunchDarkly AI Configs. The creation workflow enforces modelConfigKey presence to ensure the config and its variations are properly initialized via the API.

How can I verify that my LaunchDarkly AI Config was created correctly?▼

You can verify your LaunchDarkly AI Config by fetching it via the API. This confirms correct creation and ensures the config along with its variations are visible and accessible in the UI.

Can I attach tools to AI Config variations after creation?▼

Yes, you can attach tools to your AI Config variations. The two-step workflow creates the config first, then adds variations with models and prompts, allowing tools to be attached later in the process.