{"name":"Vidext Operator","description":"Vidext Operator is the organization-level agent exposed over MCP. It can research connected sources and organization wiki context, inspect module workflow status, and plan, build, and edit module work through a single operator agent that selects its own tools. Requests to create a course are fulfilled as modules.","iconUrl":"https://staging.vidext.com/icon.svg","category":"education","provider":{"name":"Vidext","url":"https://vidext.io"},"branding":{"name":"Vidext Operator","description":"Operate Vidext workspaces through the organization operator agent over remote MCP.","iconUrl":"https://staging.vidext.com/icon.svg","category":"education","logoUrl":"https://staging.vidext.com/apple-icon.png"},"mcp":{"transport":"streamable-http","endpoint":"https://staging.vidext.com/api/mcp/operator","protocolVersions":["2025-06-18"]},"auth":{"required":true,"type":"oauth2","authorizationServer":"https://staging.vidext.com/.well-known/oauth-authorization-server","protectedResource":"https://staging.vidext.com/.well-known/oauth-protected-resource","instructions":"OAuth 2.1 authorization-code with PKCE and dynamic client registration. No pre-provisioned tokens."},"tools":["ask_vidext_operator","list_vidext_operator_chats","read_vidext_operator_chat_history"],"instructions":"Call ask_vidext_operator with the user's request. A single operator agent handles every turn and selects its own tools. Omit chatId to create a chat or pass one to continue it; call list_vidext_operator_chats when you need a recent chatId. Before continuing a known chat whose context is unknown, call read_vidext_operator_chat_history. It returns only boundary turns; if those are insufficient, ask the operator to summarize the chat instead of attempting to recover omitted turns. When ask_vidext_operator returns operatorInputRequest, show the complete assistantText plus every question and option to the user. Continue through ask_vidext_operator with the same chatId and an open-text prompt containing either the selected option's complete prompt or the user's custom answer. The run envelope returns modelId, mutationLevel (read when only read tools ran, write when any write tool ran), availableToolNames, correlated request/run ids, and nextAction when polling is useful. Pass activeModuleId when the request is about a specific module. Every call is scoped to the authenticated user's organization membership.","serverCard":"https://staging.vidext.com/.well-known/mcp/server-card.json","documentation":"https://staging.vidext.com/llms.txt"}