LaunchDarkly MCP - Feature Flag Management for AI Agents¶
What It Is¶
LaunchDarkly MCP exposes feature flag management over the Model Context Protocol. AI agents can toggle features on/off, check flag statuses, manage progressive rollouts, run A/B experiments, and audit every flag change - bringing feature management into the agent workflow.
Tools Available¶
| Tool | Description |
|---|---|
| Flag management | List, get, toggle, and update feature flags |
| Targeting rules | Manage user segments and targeting rules |
| Experiments | Create, monitor, and conclude A/B experiments |
| Audit log | Review flag change history and who toggled what |
Quick Start¶
npx -y @launchdarkly/mcp-server
Business Use Cases¶
- Incident response: "Disable the new checkout flow feature flag in production"
- Release verification: "Is the dark mode flag at 100% rollout for all users?"
- Experiment analysis: "What's the conversion rate for the new onboarding experiment?"
Integration with CorpusIQ¶
CorpusIQ is the connector layer that lets AI assistants read the systems a business already runs on - accounting, payments, CRM, email, files and analytics - and answer with the source named. It is read-only by design: it retrieves data and never writes to, or acts in, a connected system, which is what makes it safe to point at live accounts.
This page is part of the CorpusIQ directory, the reference we maintain for the MCP server and agent-skill ecosystem. Run alongside CorpusIQ, a third-party server like this one can be combined with a company's own business data in a single answer.
Limitations¶
- LaunchDarkly account and API key required. Flag toggles in production should still follow change management processes.