LaunchDarkly MCP. Manage flags, environments, and deployments via conversation.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
LaunchDarkly MCP Server lets your AI client manage feature flags, environments, and deployments. It connects directly to your LaunchDarkly workspace, allowing you to inspect flags, list environments, and check metrics using natural conversation.
Stop leaving the platform UI to manage releases; use your agent to get real-time status updates and audit logs.
What your AI agents can do
Get environment
Gets details for a specific environment.
Get feature flag
Gets in-depth specifics for a feature flag.
Get metric
Gets details for a specific metric.
Retrieve detailed information about a specific feature flag using get_feature_flag.
List all available environments within a project using list_environments.
Retrieve a full list of account audit log entries using list_audit_logs.
Get specific information about a LaunchDarkly project using get_project.
Pull key performance metrics from experimentation using get_metric.
Get a complete list of all feature flags within a designated project using list_feature_flags.
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
019d75c5get environment
Gets details for a specific environment.
019d75c5get feature flag
Gets in-depth specifics for a feature flag.
019d75c5get metric
Gets details for a specific metric.
019d75c5get project
Gets details for a specific project.
019d75c5list audit logs
Retrieves account audit log entries for the account.
019d75c5list environments
Retrieves all environments within a project (e.g., Test, Production).
019d75c5list feature flags
Retrieves feature flags within a project.
019d75c5list metrics
Retrieves experimentation metrics within a project.
019d75c5list projects
Retrieves a list of LaunchDarkly projects.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with LaunchDarkly, then connect any of our 4,700+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,700+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
What you can do with this MCP connector
LaunchDarkly MCP Server lets your AI client manage feature flags, environments, and deployments right where you are. You don't have to jump into the LaunchDarkly UI just to check something or roll out a change. Your agent hooks directly into your workspace, letting you check flag status, list environments, and grab metrics using natural talk.
You can get real-time status updates and audit logs without lifting a finger.
How LaunchDarkly MCP Works
- 1 Install the MCP Server locally and provide your LaunchDarkly API token key.
- 2 Your AI client authenticates against the server, establishing connection to the LaunchDarkly API.
- 3 You invoke a tool (e.g.,
list_environments) via natural language, and the agent executes the API call, returning the structured data.
The bottom line is, your agent runs the LaunchDarkly commands for you, pulling data into your current chat session.
Who Is LaunchDarkly MCP For?
The DevOps engineer who needs to confirm a flag is live in Production without opening a browser. The product manager who needs immediate A/B test status checks. Fullstack developers who need to validate a new feature flag deployment quickly. If you spend time clicking through dashboards, you need this.
Runs release monitoring checks and validates environments using tools like list_environments.
Checks the status of A/B testing flags and tracks user engagement metrics using get_metric.
Verifies that a newly pushed feature flag is active in the correct environment by calling get_feature_flag.
What Changes When You Connect
- Check flag status instantly. Instead of navigating to a flag's page, use the agent to call
get_feature_flagand get its current state and targeting rules in one message. - Audit every change. Need to know who toggled a flag and when? Use
list_audit_logsto pull a complete history of account activity without digging through separate audit dashboards. - Compare environments easily. Use
list_environmentsto list all available workspaces (Staging, Production, etc.), then useget_environmentto check specific variable mappings for any of them. - Track performance data. Don't guess if an experiment is working. Call
list_metricsthenget_metricto pull specific, actionable data points directly into your chat context. - Simplify project visibility. Instead of manually browsing, use
list_projectsto see all connected workspaces andget_projectfor specific setup details. - Control the deployment process. Your agent handles the complex sequence of calls, allowing you to trigger deployments or check project status without writing CLI scripts.
Real-World Use Cases
Validating a Hotfix Flag in Production
A fullstack developer needs to confirm a hotfix flag is active only for internal users in Production. They ask their agent, which calls list_environments to confirm 'Production' exists, then calls get_feature_flag specifying 'Production' to confirm the flag status and targeting parameters.
Reviewing Recent Account Changes
The ops engineer notices suspicious activity. They ask their agent to run list_audit_logs. The agent pulls the recent log entries, letting the engineer see who accessed which project and when, solving the need for manual log review.
Comparing Staging vs. Production Variables
A product manager needs to ensure the staging environment variables match production before a release. They ask their agent to run list_environments to confirm both exist, then use get_environment to compare specific context mappings for both.
Troubleshooting Low Experiment Engagement
The product team sees low adoption rates. They ask their agent to run list_metrics to see available experiments, then call get_metric on the specific experiment ID to pull the raw data and identify the root cause.
The Tradeoffs
Assuming all flags are active
A dev assumes a flag is live because it exists in the UI, but it might be disabled for the current workspace or target group.
→
Don't trust the UI state. Always confirm the status by calling get_feature_flag, ensuring you specify the target environment and scope. This gives the ground truth state.
Ignoring the environment context
Attempting to check a flag's status without specifying if you mean 'Staging' or 'Production' leads to ambiguous results.
→
First, use list_environments to see all available workspaces. Then, always include the desired environment name when calling get_feature_flag or get_environment.
Chaining tools manually in the CLI
Writing a complex script that calls get_project then list_feature_flags then get_metric in a fixed order, which breaks if any single API endpoint changes.
→
Let your agent manage the sequence. Ask your agent to 'Give me the status of the dark mode flag in Production'—it handles the necessary calls (list_environments -> get_feature_flag -> get_metric) for you.
When It Fits, When It Doesn't
Use this MCP Server if your job involves managing feature rollouts across multiple, distinct environments (Dev, Staging, Production). You need to validate the state of a flag, check environment variables, or pull performance metrics without opening a web browser.
Don't use this if you only need to view static project metadata or if your core problem is user authentication. For pure user data lookups, a dedicated user profile tool is better. If you only need to list flags without checking their live state, the list_feature_flags tool works, but you'll lose the critical context of the environment.
This server excels at validating the state of the system, not just listing its parts.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by LaunchDarkly. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 9 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Checking flag status used to mean opening the platform, navigating to the project, selecting the environment, and finding the flag.
Before this, checking a feature flag's status was a multi-step dance. You'd click the main dashboard, find the project, select the environment (like 'Staging'), then click the flag name. If you needed to compare two flags, you'd repeat the whole process, copying and pasting statuses between tabs.
Now, you just talk to your agent. You tell it, 'What's the status of the checkout button flag in Production?' and it runs the necessary calls (`get_feature_flag`, `get_environment`) and returns the exact status, targeting percentage, and effective date in plain text. It's done.
LaunchDarkly MCP Server: Get the full context on flags and deployments.
You no longer need to juggle multiple dashboards to validate a release. The agent pulls together data from different sources—the flag status, the current environment variables, and the associated metrics—all in one chat window. This saves minutes of clicking and context switching every time you push code.
The server lets you validate the entire deployment chain, from listing the core `list_projects` to checking the final performance via `get_metric`. It keeps the entire workflow contained in your terminal or chat.
Common Questions About LaunchDarkly MCP
How do I check if a feature flag is active in Production using LaunchDarkly MCP Server? +
You ask your agent to check the flag status, specifying 'Production' as the target environment. The agent uses get_feature_flag to confirm the flag's current state and targeting percentage.
Can I list all my environments with LaunchDarkly MCP Server? +
Yes, use the list_environments tool. It retrieves all environments tied to your project (e.g., Test, Production, Beta Testing Node), so you know what workspaces are available.
What is the best way to see the history of changes with LaunchDarkly MCP Server? +
Use list_audit_logs. This tool retrieves a complete history of actions taken on the account, letting you track who changed what and when.
Does LaunchDarkly MCP Server help with A/B testing metrics? +
Yes. The list_metrics tool pulls available experimentation metrics, and get_metric lets you pull the actual data for a specific experiment to see if it's performing.
How do I find details about a specific project with LaunchDarkly MCP Server? +
Use get_project with the project ID. It fetches all core metadata about the project, giving you the full scope of the workspace.
How do I list all feature flags using the `list_feature_flags` tool? +
The list_feature_flags tool retrieves every flag configured for a project. You can use this to audit a project's full scope of feature flags and check for unused or outdated settings.
What does the `get_project` tool provide about a specific LaunchDarkly project? +
The get_project tool gives you core details about a specific project. This includes the project's name, status, and overall configuration parameters you need to manage it.
Can the LaunchDarkly MCP Server manage multiple environments using `list_environments`? +
Yes, list_environments retrieves all connected environments (like Staging, Production, etc.) within a project. This lets your agent map contexts across different deployment stages.
How do I authenticate? +
Generate a valid API access token inside your LaunchDarkly Authorization Settings and attach it.
Can I toggle flags in Production directly? +
Yes! Due to the direct integration, whatever environments the API token has rights to, the agent can configure.
Do I need to re-sync manual changes? +
No, the agent fetches live configuration directly from the provider in real-time each run.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
Temporal
Monitor and manage distributed workflows in Temporal Cloud natively via your AI agent.
Better Stack
Monitor uptime and incidents via Better Stack — list monitors, heartbeats, and on-call schedules directly from any AI agent.
Cypress Cloud
Audit E2E testing via Cypress — monitor test runs, inspect spec instances, track flaky tests, and generate enterprise reports directly from any AI agent.
You might also like
Weiban Assistant
Customer relationship management for WeCom — manage leads, customers, and group chats via AI.
Pinterest Ads
Equip your AI agent with direct access to Pinterest Ads — manage campaigns, track pin performance, and optimize shopping ad spend without opening Pinterest Ads Manager.
Zeabur (PaaS Deployment)
Deploy services, manage containers, and send transactional emails via Zeabur PaaS directly from your AI agent.