generated: '2026-08-12'
method: searched
status: announced-beta
source: https://www.sprinklr.com/newsroom/sprinklr-introduces-new-ai-capabilities-to-help-brands-move-from-insights-to-real-time-customer-action/
announced: '2026-07-15'
release: Summer '26 (26.7)
server:
  name: Sprinklr MCP
  transport: null
  url: null
  publicly_documented: false
description: 'Sprinklr announced Sprinklr MCP (Beta) in its own newsroom on 2026-07-15 as part of the
  Summer ''26 (26.7) release. The provider''s exact wording: "With Sprinklr MCP (Beta), organizations
  can access Sprinklr insights directly within AI assistants such as Microsoft Copilot, ChatGPT, and Claude
  — making customer intelligence available wherever work happens." This is a first-party Sprinklr-hosted
  MCP surface, distinct from the separate, already-shipped Model Context Protocol Tool feature that lets
  Sprinklr CONSUME external MCP servers as tools inside its Conversational AI applications.'
verified:
  provider_announcement: true
  public_endpoint: false
  public_tool_list: false
  live_tools_list_probe: false
tools: []
tools_note: NO TOOL LIST IS ASSERTED. Sprinklr publishes no MCP endpoint URL, no tools/list manifest,
  no transport declaration and no auth/scope documentation for Sprinklr MCP. There is nothing on dev.sprinklr.com
  about it — the developer portal's menu tree (35 pages) has no MCP page. Deriving a candidate tool set
  from an OpenAPI is not possible either, because Sprinklr publishes no OpenAPI. Any tool list here would
  be invented, so none is recorded.
probes:
- url: https://mcp.sprinklr.com
  status: 0
  result: DNS does not resolve
- url: https://api3.sprinklr.com/mcp
  status: 404
- url: https://www.sprinklr.com/mcp
  status: 404
- url: https://dev.sprinklr.com/mcp
  status: 200
  result: SPA catch-all HTML shell, not an MCP endpoint
- url: https://api3.sprinklr.com/mcp (POST tools/list)
  status: 404
distinct_feature:
  name: Model Context Protocol Tool
  direction: sprinklr-as-mcp-client
  docs: https://www.sprinklr.com/help/articles/integration-tools/model-context-protocol-tool/6889c27b6c75fc514a03195b
  description: Configuration surface inside the Sprinklr platform (All Settings > APIs & Integrations
    > Model Context Protocol) where an admin registers an EXTERNAL MCP server by name, authentication
    credential and endpoint URL, so its tools can be called by Sprinklr Conversational AI applications.
    This is Sprinklr acting as an MCP client, and is not the Sprinklr MCP server above.
third_party:
- name: sprinklr-mcp
  url: https://github.com/daiict218/sprinklr-mcp
  npm: sprinklr-mcp@0.1.0
  published: '2026-04-06'
  official: false
  note: Community wrapper over the Sprinklr REST API. Not endorsed or published by Sprinklr.
next_step: 'Re-probe once Sprinklr documents the endpoint. When a URL appears, POST {"jsonrpc":"2.0","id":1,"method":"tools/list"}
  with Accept: application/json, text/event-stream to capture the real inputSchemas, and only then emit
  mcp/sprinklr-tool-crosswalk.yml.'
deployment:
  mode: unclear
  verified: searched
  checked: '2026-08-12'
  source: catalog MCP census