The full upstream README, mirrored here for reference. Install config, tool schemas, adoption signals, and an original overview live on the Vidalytics MCP listing page.
One-command setup that connects your AI coding assistant to Vidalytics video analytics data via the Model Context Protocol.
Works with Claude (CLI & Desktop), Windsurf, Cursor, and any other MCP-compatible client.
That's it. The installer detects which AI clients you have installed, lets you pick which ones to configure, and wires them up. Restart the client — a browser window will open for OAuth authorization on first use.
Or via the installer:
Or manually — add this to ~/.cursor/mcp.json:
Restart Cursor. On first use a browser window opens for OAuth authorization with your Vidalytics account — no API key or environment variables to set.
$PATH, and app bundles (e.g. /Applications on macOS)--client--client, --all, --yes) skip the checklist and behave predictablyOnce connected, your AI assistant gains access to:
| Tool | Description |
|---|---|
set_user_context | MUST be called before any other tool to enable analytics |
list_videos | List videos with pagination |
get_video | Get video details |
get_video_by_embed_guid | Find a video by its embed GUID |
update_video | Update a video's title or folder |
get_video_embed | Get the embed code and configuration |
get_video_settings | Get playback settings (autoplay, controls, etc) |
get_video_thumbnail | Get the thumbnail image URL |
get_video_stats | Views, play rate, watch time, conversions |
get_video_dropoff | Audience retention by percentage |
get_video_percentage_watched | % of viewers who reached each point |
get_video_live_metrics | Real-time active viewers and watch rate |
get_video_ctas | Get CTAs for a video |
get_video_pause_screens | Get pause screens for a video |
get_videos_stats_batch | Stats for up to 30 videos at once |
get_videos_timeline | Timeline stats for up to 5 videos |
list_folders | List video folders |
list_settings_templates | List settings templates |
get_api_usage | Get current API usage and quota |
list_connections | List apps connected to your account |
revoke_connection | Disconnect an app or yourself |
upload_video_from_url | Upload a video from a remote URL |
get_video_upload_url | Get a signed URL for local file upload |
validate_upload | Complete a direct video upload |
publish_video | Publish a video's pending draft settings |
duplicate_video | Duplicate a video and publish the copy |
create_folder | Create a video folder, optionally nested under another folder |
apply_settings_template | Apply a reusable player settings template to a video |
create_video_cta | Create a call-to-action on a video |
update_video_cta | Update an existing call-to-action on a video |
set_video_thumbnail_from_url | Set a video's thumbnail from a public image URL |
set_video_thumbnail_from_frame | Set a video's thumbnail from one of its frames |
delete_video_thumbnail | Remove a custom thumbnail and restore the default |
Run with no flags in an interactive terminal to get a checklist of clients to configure (detected ones are pre-selected; use space to toggle, enter to confirm). --client cursor,windsurf does the same selection non-interactively.
The --config flag can be repeated for multiple files. The target file must follow the { "mcpServers": {} } format used by Claude Desktop, Cursor, and Windsurf — useful for unsupported clients like Zed or VS Code with an MCP plugin.
Authorization issues, or need to re-authenticate? Reset the credentials that
mcp-remote caches in your home directory, then restart the client:
| OS | Command |
|---|---|
| macOS / Linux | rm -rf ~/.mcp-auth |
| Windows (CMD) | rd /s /q "%USERPROFILE%\.mcp-auth" |
| Windows (PowerShell) | Remove-Item -Recurse -Force "$HOME\.mcp-auth" |
This server is published to the official MCP Registry as com.vidalytics/mcp. It is a remote (streamable-http) server, so registry-aware MCP clients can connect to it directly at:
No API key or environment variables are required — authorization is handled via OAuth on first use.