What it does

Reads and changes feature flags through the LaunchDarkly API, so an agent can turn a flag off during an incident instead of asking a human to open a dashboard. The risk is obvious and real: a mistaken write flips behavior for production users. Give it a non-production environment token and promote changes by hand.

Install

Terminal

npx -y @launchdarkly/mcp-server

Client configuration

Paste this into your MCP client config (Claude Desktop, Cursor, or any client that reads mcpServers), then restart the client.

mcp.json

{
  "mcpServers": {
    "launchdarkly": {
      "command": "npx",
      "args": [
        "-y",
        "@launchdarkly/mcp-server"
      ]
    }
  }
}

Use it for

Toggling flags and reading rollout state while shipping a change

Skip it if

Teams without a LaunchDarkly account, or anyone who should not hold flag-write access

When to use it

Reach for this when the task needs feature-flags inside the agent loop. Skip it if a plain shell command or a one-off script is enough — every server you add costs context and adds a failure mode.

Other TypeScript servers

Filesystem — Gives an agent scoped read, write, and search access to directories you explicitly allow
Memory — A persistent knowledge graph stored in a local JSON file, exposed through entities, relations, and observations
Sequential Thinking — Gives the model a scratchpad for structured, revisable reasoning: numbered thoughts that can branch, revise, and conclude

LaunchDarkly MCP