All skills
launchdarkly avatar

/projects

@ef54971 official

Guide for setting up LaunchDarkly projects in your codebase. Helps you assess your stack, choose the right approach, and integrate project management that makes sense for your architecture.

Use this Skill: https://skilld.dev/gh/launchdarkly/agent-skills/projects

This session only. Nothing lands on disk.

referencesquick-start.md

≈921 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Quick Start: Create Your First Project

Basic project creation patterns for getting started quickly.

Prerequisites

  • LaunchDarkly API access token with projects:write permission
  • Set LAUNCHDARKLY_API_TOKEN environment variable

Basic Project Creation

Using the LaunchDarkly API

Endpoint: POST https://app.launchdarkly.com/api/v2/projects

Required Headers:

Authorization: {YOUR_API_TOKEN}
Content-Type: application/json

Minimal Payload:

{
  "name": "My Agent Project",
  "key": "my-ai-project"
}

Recommended Payload:

{
  "name": "Customer Support Agent",
  "key": "support-ai",
  "tags": ["ai-configs", "production"]
}

Response Handling

Success (201 Created)

{
  "name": "Customer Support Agent",
  "key": "support-ai",
  "environments": {
    "items": [
      {
        "name": "Production",
        "key": "production",
        "apiKey": "sdk-xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
      },
      {
        "name": "Test",
        "key": "test",
        "apiKey": "sdk-yyyyyyyy-yyyy-yyyy-yyyy-yyyyyyyyyyyy"
      }
    ]
  }
}

Action: Extract and save SDK keys for use in your application.

Conflict (409)

Project with that key already exists.

Action: Either use the existing project or choose a different key.

Error (400, 401, 403)

{
  "code": "invalid_request",
  "message": "project key must be lowercase with hyphens"
}

Action: Fix the payload based on error message.

Project Key Rules

Must follow these constraints:

  • Pattern: ^[a-z][a-z0-9-]*$
  • Start with: Lowercase letter
  • Contains only: Lowercase letters, numbers, hyphens
  • Unique: Across your entire organization

Valid Examples

support-ai
chat-bot-v2
recommendation-engine
customer-ai-prod

Invalid Examples

Support_AI          # uppercase and underscore
123-project         # starts with number
my.project          # contains dot
ai_chatbot          # underscore not allowed

Extracting SDK Keys

After creating a project, you'll need the SDK keys to connect your application.

Environments Created by Default

  • Production (key: production)
  • Test (key: test)

Get SDK Key for an Environment

Endpoint: GET https://app.launchdarkly.com/api/v2/projects/{projectKey}?expand=environments

Parse Response:

{
  "environments": {
    "items": [
      {
        "key": "production",
        "apiKey": "sdk-xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
      }
    ]
  }
}

Filter by environment.key to find the desired environment's apiKey.

List Existing Projects

Before creating a new project, you may want to check what exists.

Endpoint: GET https://app.launchdarkly.com/api/v2/projects

Response:

{
  "items": [
    {
      "name": "Customer Support Agent",
      "key": "support-ai",
      "tags": ["ai-configs"]
    }
  ]
}

Common Mistakes

Mistake Problem Fix
Uppercase in key Rejected by API Use lowercase only
Spaces in key Invalid format Use hyphens instead
Key collision 409 conflict Choose unique key or use existing
Missing API token 401 unauthorized Set LAUNCHDARKLY_API_TOKEN
Wrong permission 403 forbidden Request projects:write permission

Next Steps

After creating your project:

  1. Save SDK keys to your environment configuration
  2. Initialize LaunchDarkly SDK in your application
  3. Create configs within the project
  4. Test the integration in test environment first

See Environment Configuration for saving SDK keys to your codebase.

Source: SKILL.md on GitHub

2 alerts2d3 checks · Risk HIGH
  • Gen Agent Trust Hub2d

    The skill facilitates LaunchDarkly project setup but performs high-risk operations, including instructions to automatically extract credentials from sensitive agent configuration files (~/.claude/config.json) and system environment variables. It also generates and executes local scripts to verify integrations, creating a potential path for local code execution.

  • Socket2d

    1 alert: gptAnomaly

  • Snyk2d

    Risk: HIGH · 1 issue

Signed by skilld at ef54971. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 2 days ago.

Activeupdated 3 days ago
metadata
{
  "author": "launchdarkly",
  "version": "0.4.0"
}
Other metadata
compatibility
Requires LaunchDarkly API access token with projects:write permission or LaunchDarkly MCP server.
  • Python
  • Go
  • Infrastructure
  • launchdarkly
  • project-setup
  • api-integration
  • nodejs
  • configuration-management
  • feature-flags
  • secrets-management

README badge

README badge for launchdarkly/agent-skills/projects

Guides setup of LaunchDarkly projects in a codebase by exploring the stack, assessing architecture, and choosing the right implementation path. Supports Python, Node.js, Go, and polyglot setups, with reference guides for quick setup, environment configuration, project cloning, and infrastructure automation.

Generated from the current SKILL.md.

What permissions do I need to set up LaunchDarkly projects?
Your API access token must have the projects:write permission, or you can use a LaunchDarkly MCP server configured in your environment instead.
Can the skill automatically detect my LaunchDarkly API key?
Yes. The skill checks for LAUNCHDARKLY_API_KEY, LAUNCHDARKLY_API_TOKEN, or LD_API_KEY environment variables, and also checks the Claude MCP config at ~/.claude/config.json before prompting you for the key.
What project key format is required?
Project keys must start with a letter, use only lowercase letters, numbers, and hyphens, and cannot contain uppercase letters, underscores, or dots.
Does this skill create the projects, or just guide me?
The skill guides you through the setup process and follows reference implementations for your tech stack (Python, Node.js, Go, etc.), but you choose which reference path to execute based on your architecture.
What should I do after projects are created?
After setup, verify the project exists and SDK keys are valid, then proceed to the configs-create skill to create feature flags or the sdk skill to integrate into your application.

Generated from the current SKILL.md. These answers refresh after source changes.