Skip to content

MCP Server

The Framedash MCP Server is a Model Context Protocol server that gives LLMs direct access to Framedash telemetry data. It provides 12 read-only tools and 4 resources, enabling natural-language queries against your game analytics.

Add to your claude_desktop_config.json:

{
"mcpServers": {
"framedash": {
"command": "npx",
"args": ["-y", "@framedash/mcp-server"],
"env": {
"FRAMEDASH_API_KEY": "fd_admin_xxx",
"FRAMEDASH_PROJECT_ID": "your-project-uuid"
}
}
}
}

Add to your VS Code settings:

{
"claude.mcpServers": {
"framedash": {
"command": "npx",
"args": ["-y", "@framedash/mcp-server"],
"env": {
"FRAMEDASH_API_KEY": "fd_admin_xxx",
"FRAMEDASH_PROJECT_ID": "your-project-uuid"
}
}
}
}
VariableRequiredDescription
FRAMEDASH_API_KEYYesAdmin API key (fd_admin_ prefix)
FRAMEDASH_PROJECT_IDNoDefault project UUID
FRAMEDASH_BASE_URLNoAPI base URL (default: https://app.framedash.dev)
ToolDescriptionArguments
queryExecute a read-only SQL query against the ClickHouse events tablesql (string, required), project_id (uuid, optional), limit (int 1-1000, default 100)
ToolDescriptionArguments
get_dashboardProject KPIs (DAU, MAU, sessions, events)project_id (uuid, optional), days (7/14/30/90, default 30)
get_retentionCohort retention analysisproject_id (uuid, optional), days (7/14/30/90, default 30)
get_funnelEvent funnel analysisproject_id (uuid, optional), steps (string, required: comma-separated 2-8 event names), days (7/14/30/90, default 30)
get_insightsAggregated insights by dimensionproject_id (uuid, optional), metric (count/unique_players, required), group_by (string, required: event_name, platform, etc.), days (7/14/30/90, default 30), limit (10/20/50), event_name (string, optional)
get_heatmapHeatmap grid data for a mapproject_id (uuid, optional), map_id (string, required), cell_size (5/10/25/50, default 25), days (1/7/14/30, default 7), event_name (string, optional)
ToolDescriptionArguments
list_projectsShow projects bound to the API key(none)
get_project_statusProject health overview (event counts, last event time)project_id (uuid, optional)
list_mapsList maps in the projectproject_id (uuid, optional)
list_contentList content registry entriesproject_id (uuid, optional), type (string, optional)
ToolDescriptionArguments
list_alertsList alert rulesproject_id (uuid, optional)
get_alert_historyAlert trigger/resolve historyproject_id (uuid, optional), limit (int 1-100, default 50)

Data is also available as MCP resources via the framedash:// URI scheme:

URIDescription
framedash://projectsProjects bound to the API key
framedash://projects/{projectId}/mapsMaps with coordinates and bounds
framedash://projects/{projectId}/contentContent registry entries
framedash://projects/{projectId}/statusProject health and statistics

Once the MCP Server is configured, you can query your data in natural language:

PromptTool Used
”Show me DAU for the last 7 days”get_dashboard (days=7)
“Analyze the funnel from spawn to death”get_funnel (steps=“player.spawn,player.death”)
“Break down event counts by platform”get_insights (metric=count, group_by=platform)
“Show the FPS heatmap for my map”get_heatmap (map_id=…)
”Check recent alert history”get_alert_history