@featureflip/mcp
MCP (Model Context Protocol) server for Featureflip — manage feature flags from AI coding assistants (Claude Code, Cursor, Copilot, Cline) and autonomous agents.
Setup
- Create an API token in the Featureflip dashboard:
- Personal token (
ffp_...): Settings → API Tokens — acts as you, for interactive editor use. - Service token (
ffs_...): Organization Settings → Service Tokens — scoped machine identity for CI/agents.
- Personal token (
- Add the server to your MCP client:
Claude Code
claude mcp add featureflip -e FEATUREFLIP_TOKEN=ffp_your_token -- npx -y @featureflip/mcp@latest
Cursor / Cline / generic JSON config
{
"mcpServers": {
"featureflip": {
"command": "npx",
"args": ["-y", "@featureflip/mcp@latest"],
"env": { "FEATUREFLIP_TOKEN": "ffp_your_token" }
}
}
}
Keep the @latest. Without a version, npx reuses the copy it downloaded the first time and never
checks npm again, so tools added in later releases would never show up in your client.
Configuration
| Env var | Required | Default | Purpose |
|---|---|---|---|
FEATUREFLIP_TOKEN | yes | — | ffp_/ffs_ API token |
FEATUREFLIP_API_URL | no | https://api.featureflip.io | API base URL |
FEATUREFLIP_ORG | no | auto | Org slug (needed only for multi-org personal tokens) |
Tools
CRUD: list_projects, list_environments, list_flags, get_flag, create_flag, update_flag,
delete_flag, archive_flag, restore_flag, set_flag_expiry, set_flag_owner, set_flag_permanent,
toggle_flag, update_flag_environment_config, copy_flag_config, get_targeting, update_targeting,
manage_targeting_rule, manage_variation, list_segments, get_segment, manage_segment,
manage_project, set_default_flag_lifetime, manage_environment, list_sdk_keys, manage_sdk_key,
list_flag_triggers, manage_flag_trigger, list_rollout_ramps, manage_rollout_ramp
Webhooks (Admin token required): list_webhooks, list_webhook_deliveries, manage_webhook, deliver_webhook
Workflows: flag_status (cross-environment view), flag_evaluations (what a flag served in one environment: evaluation counts by variation and rule), compare_environments (which flags differ between two environments),
find_stale_flags (cleanup candidates),
list_removal_candidates (flags safe to remove, with the prerequisite dependents to remove first),
wrap_feature (create flag + get the SDK snippet for your language)
Full reference: https://featureflip.io/docs/integrations/mcp/
Notes
- Flag evaluation is not exposed over MCP — use the language SDKs in application code.
- Every mutation is audit-logged and attributed to the token.
License
Apache-2.0