Mindstamp MCP

Mindstamp exposes a Model Context Protocol (MCP) server so AI agents can operate
a Mindstamp account: build and edit interactive video, change settings, and read
viewer and results data. This is the developer reference for connecting.

• Endpoint: https://app.mindstamp.com/mcp (Streamable HTTP, JSON-RPC 2.0)
• Auth: OAuth 2.1 (authorization code + PKCE) with automatic discovery, or an
  existing Mindstamp API token as a bearer token
• Availability: read access on paid plans; write access on Pro and Enterprise
• AI features are off until an account owner opts in

Connect

Any MCP client with OAuth (Claude, ChatGPT, Cursor)

Point the client at the endpoint and authorize once. Clients that support
dynamic client registration and OAuth discovery need no manual setup:

https://app.mindstamp.com/mcp

The client discovers the authorization server from
https://app.mindstamp.com/.well-known/oauth-protected-resource, registers
itself at /oauth/register, and runs the standard code + PKCE flow. Access
tokens are short-lived and refresh silently; you are not re-prompted.

Claude Code (CLI)

claude mcp add --transport http mindstamp https://app.mindstamp.com/mcp

With an existing API token

Send your Mindstamp API token as a bearer token:

Authorization: Bearer <your-api-token>

Create tokens in your Mindstamp account (API access is a Pro/Enterprise feature).

Scopes

An existing API token (scope public) has full access, matching the REST API.
Tools that need viewer identities require read:viewers; without it, viewer
data is returned with identities redacted. Tier‑3 tools (delete, publish,
account brand) require an explicit confirm: true after human approval.

Tools

Read: list_videos, get_video, get_transcript, get_video_stats,
get_question_results, list_views, get_view, list_interactions.

Build and edit: add_button, add_questions, add_text, add_hotspot,
set_chapters, delete_interaction.

Settings: update_video_settings, set_end_screen_link,
set_end_screen_branch, configure_lead_capture.

Account and high-impact: apply_brand_design, set_publish_state,
delete_video (each requires confirm: true).

Call tools/list for the current set and each tool's input schema.

Example

// tools/call — add a knowledge check at 45 seconds
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "add_questions",
    "arguments": {
      "video_id": "<uuid>",
      "questions": [{
        "prompt": "Which fire class involves flammable liquids?",
        "answers": ["Class A", "Class B", "Class C", "Class K"],
        "correct_answer": "Class B",
        "time": 45
      }]
    }
  }
}

Safety

• Every action is authorized as the connecting user, within their permissions.
• Every action is recorded in the account's activity log.
• Reversible (tier‑2) changes can be undone inside Mindstamp.
• Destructive, publishing, and brand actions require confirmation.
• Product data (transcripts, titles, viewer answers) is treated as data, never
  as instructions to the agent.

Discovery documents

• https://app.mindstamp.com/.well-known/oauth-protected-resource
• https://app.mindstamp.com/.well-known/oauth-authorization-server
• https://app.mindstamp.com/.well-known/mcp/server-card.json
• https://app.mindstamp.com/auth.md