Connect
Optimize
Secure
LaunchDarkly MCP Server
for AI Agents
Production-ready LaunchDarkly MCP server with 85 extensible actions — plus built-in authentication, security, and optimized execution.
LaunchDarkly MCP Server
Built by 
StackOne
StackOne 
Coverage
85 Agent Actions
Create, read, update, and delete across LaunchDarkly — and extend your agent's capabilities with custom actions.
Authentication
Agent Tool Authentication
Per-user OAuth in one call. Your LaunchDarkly MCP server gets session-scoped tokens with zero credentials stored on your infra.
Agent Auth →
Security
Agent Protection
Every LaunchDarkly tool response scanned for prompt injection in milliseconds — 88.7% accuracy, all running on CPU.
Prompt Injection Defense →
Performance
Max Agent Context. Min Cost.
Free up to 96% of your agent's context window to enhance reasoning and reduce cost, on every LaunchDarkly call.
Tools Discovery →What is the LaunchDarkly MCP Server?
A LaunchDarkly MCP server lets AI agents read and write LaunchDarkly data through the Model Context Protocol — Anthropic's open standard for connecting LLMs to external tools. StackOne's LaunchDarkly MCP server ships with 85 pre-built actions, fully extensible via the Connector Builder — plus managed authentication, prompt injection defense, and optimized agent context. Connect it from MCP clients like Claude Desktop, Cursor, and VS Code, or from agent frameworks like OpenAI Agents SDK, LangChain, and Vercel AI SDK.
All LaunchDarkly MCP Tools and Actions
Every action from LaunchDarkly's API, ready for your agent. Create, read, update, and delete — scoped to exactly what you need.
Projects
- Create Project
Create a new project
- List Projects
Retrieve all projects in the LaunchDarkly account
- Get Project
Retrieve a specific project by key
- Update Project
Update a project using JSON Patch
- Delete Project
Delete a project
Feature Flags
- Create Feature Flag
Create a new feature flag
- Copy Feature Flag
Copy a feature flag to a new flag
- List Feature Flags
Retrieve all feature flags in a project
- Get Feature Flag
Retrieve a specific feature flag by key
- Update Feature Flag
Update a feature flag using JSON Patch
- Delete Feature Flag
Delete a feature flag
Environments
- Create Environment
Create a new environment in a project
- List Environments
Retrieve all environments in a project
- Get Environment
Retrieve a specific environment by key
- Update Environment
Update an environment using JSON Patch
- Delete Environment
Delete an environment
Segments
- Create Segment
Create a new segment
- List Segments
Retrieve all segments in a project and environment
- Get Segment
Retrieve a specific segment by key
- Update Segment
Update a segment using JSON Patch
- Delete Segment
Delete a segment
Account Members
- List Account Members
Retrieve all account members
- Get Account Member
Retrieve a specific account member by ID
- Update Account Member
Update an account member using JSON Patch
- Delete Account Member
Remove a member from the account
Teams
- Create Team
Create a new team
- List Teams
Retrieve all teams
- Get Team
Retrieve a specific team by key
- Update Team
Update a team using instructions
- Delete Team
Delete a team
Custom Roles
- Create Custom Role
Create a new custom role
- List Custom Roles
Retrieve all custom roles
- Get Custom Role
Retrieve a specific custom role by key
- Update Custom Role
Update a custom role using JSON Patch
- Delete Custom Role
Delete a custom role
Experiments
- Create Experiment
Create a new experiment
- List Experiments
Retrieve all experiments in a project and environment
- Get Experiment
Retrieve a specific experiment by key
- Update Experiment
Update an experiment using instructions
Metrics
- Create Metric
Create a new metric
- List Metrics
Retrieve all metrics in a project
- Get Metric
Retrieve a specific metric by key
- Update Metric
Update a metric using JSON Patch
- Delete Metric
Delete a metric
Contexts
- Get Context
Retrieve a specific context by kind and key
- Search Contexts
Search for contexts across all kinds
Context Instances
- Search Context Instances
Search for context instances
- Delete Context Instance
Delete a context instance by ID
Context Kinds
- Create Context Kind
Create a new context kind
- List Context Kinds
Retrieve all context kinds in a project
- Get Context Kind
Retrieve a specific context kind by key
- Update Context Kind
Update a context kind using JSON Patch
Expiring Context Targets
- Get Expiring Context Targets
Retrieve expiring context target assignments for a flag
- Update Expiring Context Targets
Update expiring context target assignments for a flag
Webhooks
- Create Webhook
Create a new webhook
- List Webhooks
Retrieve all webhooks
- Get Webhook
Retrieve a specific webhook by ID
- Update Webhook
Update a webhook using JSON Patch
- Delete Webhook
Delete a webhook
Integration Subscriptions
- Create Integration Subscription
Create a new integration subscription
- List Integration Subscriptions
Retrieve all integration subscriptions
- Get Integration Subscription
Retrieve a specific integration subscription by ID
- Update Integration Subscription
Update an integration subscription using JSON Patch
- Delete Integration Subscription
Delete an integration subscription
Flag Triggers
- Create Flag Trigger
Create a new flag trigger
- List Flag Triggers
Retrieve all triggers for a feature flag
- Get Flag Trigger
Retrieve a specific flag trigger by ID
- Update Flag Trigger
Update a flag trigger using JSON Patch
- Delete Flag Trigger
Delete a flag trigger
AI Configs
- Create AI Config
Create a new AI config
- List AI Configs
Retrieve all AI configs in a project
- Get AI Config
Retrieve a specific AI config by key
- Update AI Config
Update an AI config using JSON Patch
- Delete AI Config
Delete an AI config
Model Configurations
- Create Model Configuration
Create a new AI model configuration
- List Model Configurations
Retrieve all AI model configurations in a project
- Get Model Configuration
Retrieve a specific AI model configuration by key
- Update Model Configuration
Update an AI model configuration using JSON Patch
- Delete Model Configuration
Delete an AI model configuration
Other
- Get Experiment Results
Retrieve experiment results and metrics
- Get Context Attribute Names
Retrieve attribute names for a context kind
- Get Context Attribute Values
Retrieve attribute values for a context kind and attribute
- List Audit Log Entries
Retrieve audit log entries with filtering
- Get Audit Log Entry
Retrieve a specific audit log entry by ID
- Invite Account Member
Invite a new member to the account
Set Up Your LaunchDarkly MCP Server in Minutes
One endpoint. Any framework. Your agent is talking to LaunchDarkly in under 10 lines of code.
MCP Clients
Agent Frameworks
JSON{
"mcpServers": {
"stackone": {
"command": "npx",
"args": [
"-y",
"mcp-remote@latest",
"https://api.stackone.com/mcp?x-account-id=<account_id>",
"--header",
"Authorization: Basic <YOUR_BASE64_TOKEN>"
]
}
}
}More Developer Tools MCP Servers
Azure DevOps
172+ actions
Cloudflare
137+ actions
Bitbucket
134+ actions
Supabase
128+ actions
GitLab
125+ actions
Terraform
118+ actions
OneLogin
109+ actions
LaunchDarkly MCP Server Resources
MCP Code Mode: Keeping Tool Responses Out of Agent Context
Anthropic's code_execution processes data already in context. Custom MCP code mode keeps raw tool responses in a sandbox. 14K tokens vs 500.
11 min
Comparing BM25, TF-IDF, and Hybrid Search for MCP Tool Discovery
Benchmarking BM25, TF-IDF, and hybrid search for MCP tool discovery across 916 tools. The 80/20 TF-IDF/BM25 hybrid hits 21% Top-1 accuracy in under 1ms.
10 min
Indirect Prompt Injection Defense for MCP Tools: A Technical Guide
MCP tools that read emails, CRM records, and tickets are indirect prompt injection vectors. Here's how we built a two-tier defense that scans tool results in ~11ms.
12 min
LaunchDarkly MCP Server FAQ
LaunchDarkly MCP server vs direct API integration — what's the difference?
A LaunchDarkly MCP server and direct API integration serve different use cases. Direct API integration is for software-to-software — backend code calling LaunchDarkly. A LaunchDarkly MCP server is for AI agents — MCP clients like Claude and Cursor, plus framework agents built with OpenAI, LangChain, or Vercel AI — discovering and calling LaunchDarkly at runtime. StackOne provides both.
How does LaunchDarkly authentication work for AI agents?
LaunchDarkly authentication for AI agents works through a StackOne Connect Session. Create one via the dashboard or the SDK — you get an auth link and ready-to-paste config for Claude Desktop, Cursor, and other MCP clients. Your user authenticates their own LaunchDarkly account; StackOne handles token exchange, storage, and refresh. Credentials never reach the LLM, and each user is isolated via
origin_owner_id.Are LaunchDarkly MCP tools vulnerable to prompt injection?
Yes — LaunchDarkly MCP tools can be vulnerable to indirect prompt injection. Any tool that reads user-written content — documents, messages, tickets, records, or free-text fields — is a potential vector. StackOne Defender scans every tool response before it enters the agent's context — regex patterns in ~1ms, then a MiniLM classifier in ~4ms. 88.7% accuracy, CPU-only.
What is the context bloat of a LaunchDarkly agent and how do I avoid it?
Context bloat happens when LaunchDarkly tool schemas and API responses eat your LaunchDarkly agent's memory, preventing it from reasoning effectively. A single LaunchDarkly query can return a massive JSON response, and connecting multiple tools compounds the problem. Tools Discovery and Code Mode reduce context bloat — loading only relevant tools per query and keeping raw responses out of the agent's context.
Can I limit which actions my LaunchDarkly agent can access?
Yes — you can limit which actions your LaunchDarkly agent can access directly from the StackOne dashboard. Toggle actions on or off, or restrict them to specific accounts, with no code changes to your agent. Session tokens can be scoped to exact actions so if one leaks, exposure stays contained.
Can I create custom agent actions for my LaunchDarkly MCP server?
Yes — you can create custom agent actions for your LaunchDarkly MCP server using Connector Builder. It's an integration agent your coding assistant (Claude Code, Cursor, or Copilot) can invoke to research LaunchDarkly's API, generate production-ready connector YAML, test against the live API, and validate before you ship.
When should I NOT use a LaunchDarkly MCP server?
Skip a LaunchDarkly MCP server if your integration is purely software-to-software — direct LaunchDarkly API integration is simpler when no AI agent is involved. For deterministic, compliance-critical operations (financial transactions, regulatory reporting), direct API gives you predictable behavior without agent-driven decision-making. MCP shines when AI agents need to dynamically discover and call LaunchDarkly actions at runtime.
What AI frameworks and AI clients does the StackOne LaunchDarkly MCP server support?
The StackOne LaunchDarkly MCP server supports both. MCP clients (paste-and-go apps): Claude Desktop, Claude Code, Cursor, VS Code, Goose. Agent frameworks (code SDKs you build with): OpenAI Agents SDK, Anthropic, Vercel AI, Google ADK, CrewAI, Pydantic AI, LangChain, LangGraph, Azure AI Foundry.
Put your AI agents to work
All the tools you need to build and scale AI agent integrations, with best-in-class connectivity, execution, and security.