Valossa Assistant (multimodal Video Ai Agent)
Valossa Assistant MCP server
Valossa Assistant MCP server — bring Valossa’s multimodal AI video analysis into Claude and your AI agents: ask your video library anything and get answers, soundbites, transcripts, metadata and finished clips.
New here?
You need a Valossa account to use this server. Create an account or sign in on the Valossa web app.
MCP endpoint
https://mcp-assistant.valossa.com/mcp
Transport: streamable-http. Authentication: OAuth 2.0
(the client opens a browser login on first connect).
Add it to your client
Claude Code (CLI):
claude mcp add --transport http valossa https://mcp-assistant.valossa.com/mcp
Claude Desktop, Cursor, and other JSON-config clients — add an entry like:
{
"mcpServers": {
"valossa": {
"type": "http",
"url": "https://mcp-assistant.valossa.com/mcp"
}
}
}
(Some clients call the key servers or the field
transport instead of type — check that client's
own MCP documentation. The URL and the streamable-http transport are the
same everywhere.)
Any other MCP client: point it at the endpoint above using
the streamable-HTTP transport. Clients that support MCP OAuth will handle
login automatically; the discovery metadata is at
/.well-known/mcp.json and
/.well-known/oauth-protected-resource.
After adding it, your client will open a Valossa login page the first time you connect.
Documentation
https://mcp-assistant.valossa.com/documentation
Privacy policy
https://valossa.com/privacy-policy/
Terms&Conditions
Server Config
{
"mcpServers": {
"valossa": {
"type": "http",
"url": "https://mcp-assistant.valossa.com/mcp"
}
}
}