The ConfigCat MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your ConfigCat account. It provides structured and secure access to your products, configurations, environments, feature flags, settings, segments, and tags, so your agent can inspect configurations, create flags, update environment values, manage targeting rules, and edit flag metadata on your behalf.
- Account and product discovery: Have your agent list accessible organizations and products, then review their configurations, environments, segments, and tags.
- Feature flag and setting management: Let the agent list existing flags and typed settings, create new ones, or permanently delete those you no longer need.
- Environment value inspection: Instruct your agent to review a flag's value and complete evaluation logic in a selected environment.
- Targeting and rollout updates: Direct your agent to change selected flag values or replace complete evaluation rules for a specific environment.
- Flag metadata maintenance: Have your agent update flag names, descriptions, and tag assignments while keeping other metadata unchanged.