Kinescope MCP FAQ

Answers to common questions about working with Kinescope MCP Server.

General questions

What is MCP and why is it needed?

Model Context Protocol (MCP) is an open standard developed by Anthropic that allows AI assistants to securely interact with external services and data.

Why MCP is needed for Kinescope:

• Manage videos through natural language
• Automate routine tasks
• Quickly create reports and analyze data
• integrate Kinescope with other tools through AI

Which platforms support MCP?

Kinescope MCP Server works with any platform supporting the MCP protocol. We recommend four options:

1. Cursor (for developers)

• Built-in MCP support
• Code integration
• JSON configuration setup
• Paid iDE (~$20/mo)
• Globally available
• Requires: Kinescope API token

2. Claude Desktop (recommended for general use)

• Official AI assistant from Anthropic
• Quick terminal installation
• Full out-of-the-box MCP support
• Paid subscription (~$20/mo)
• Globally available
• Requires: Kinescope API token

3. Claude Code (for developers with Claude)

• iDE with built-in Claude
• JSON configuration setup
• Code integration
• Paid iDE (~$20/mo)
• Globally available
• Requires: Kinescope API token

4. Qwen3 (desktop application)

• Desktop app with MCP support
• JSON configuration setup
• Free to use
• Requires: Kinescope API token

Recommendations:

• For developers — Cursor or Claude Code (code integration)
• For general use — Claude Desktop (official AI assistant)
• Others — any of the above platforms

See the current list of all MCP clients in the MCP documentation .

is there a cost for using MCP?

For Kinescope MCP Server:

• The MCP Server itself is provided for free
• Standard limits of your Kinescope plan apply
• Recommend the Business plan or higher for full access

For the platform:

Additional requirements:

• Kinescope API token — free (included in the plan)

is using MCP safe?

Yes, it is safe:

• All data is transmitted via secure HTTPS connection
• Your Kinescope account access rights are enforced
• Delete operations are disabled for security
• All actions are logged

Security recommendations:

• Do not share connection settings with others
• Use strong passwords for accounts
• Regularly check activity logs

Setup and connection

How do i configure the connection to MCP Server?

The current step-by-step setup for different clients (Cursor, Claude Desktop, Claude Code, Qwen3, etc.) is described in the Configuring the MCP Server connection section.

Can i use web versions of AI assistants?

No, web versions are not supported.

Why:

• Kinescope MCP Server requires specifying an API token in the configuration
• Web versions (qwen.chat, etc.) do not allow specifying API tokens
• This is a security limitation of the web platforms themselves

What to use instead:

• Cursor — iDE for developers
• Claude Desktop — official AI assistant from Anthropic
• Claude Code — iDE with built-in Claude
• Qwen3 — desktop application (free)

How to get started:

1. Download Cursor, Claude Desktop, Claude Code, or Qwen3 (desktop application)
2. Get an API token in Kinescope
3. Configure the connection following the instructions
4. Start working

Can i use free options?

Yes, there are free options:

Free options:

Additional requirements (free):

• Kinescope API token — free (included in the plan)

Recommendations:

• For development — Cursor Trial or Claude Code Trial, then paid subscription
• For general use — Claude Desktop
• For ongoing work — Cursor, Claude Desktop, or Claude Code
• For users wanting a free option — Qwen3

Why is MCP Server not responding?

Possible causes (wrong API token, network, configuration error) and step-by-step solutions for Cursor, Claude Desktop, Claude Code, and Qwen3 are described in the Troubleshooting section of the main guide.

Can i use multiple Kinescope accounts?

Yes, you can configure multiple connections:

in Cursor or Claude Code:

• Add multiple servers to the MCP configuration
• Give different names: kinescope-work, kinescope-personal

in Claude Desktop:

• Add multiple servers via the Smithery CLI (mcp add) with different names
• Manually configure for each account

in Qwen3:

• Add multiple servers in MCP settings
• Name them clearly: “Kinescope Work”, “Kinescope Personal”

Switching between accounts:

Specify which account to use in the query, or switch the active server in settings.

How do i switch between projects?

MCP Server automatically works with all projects you have access to. Simply specify the project name in the query:

Show videos from the "Marketing" project

Working with data

What data is available through MCP?

Available:

• Video information (metadata, status)
• View analytics
• Projects and folders
• Playlists
• Subtitles and chapters
• Streams and recordings
• Billing statistics

Unavailable:

• Data deletion
• Account settings
• Payment information
• Direct video file upload

Can i edit videos through MCP?

You can edit video metadata:

• Title and description
• Tags
• Privacy settings
• Chapters
• Subtitles

Cannot edit:

• The video file itself
• Trim or edit video
• Change encoding

Use the web interface or specialized tools for file editing.

How often is data updated?

Data is updated in real time. When you make a query through MCP Server, you get current information from your Kinescope account.

Note:

• Analytics may update with a small delay (up to 5 minutes)
• Processing new videos takes time

Can i download videos through MCP?

No, direct video file download through MCP is unavailable.

What is available:

• Get video links
• Export metadata
• Get subtitle and supplementary material links

Use the Kinescope web interface for downloading.

Analytics and reports

What metrics are available through MCP?

Basic metrics:

• View count
• Unique viewers
• Watch time
• Completion rate (percentage watched)

Advanced metrics:

• Retention rate (audience retention)
• View geography
• Devices and browsers
• Viewer drop-off points

Billing:

• Traffic usage
• Storage usage
• Statistics by project

Metric availability depends on your Kinescope plan.

Can i export data?

Yes, you can ask the AI to export data:

Export statistics for all videos for the month to Google Sheets

Create a CSV file with a list of all videos and their metadata

The AI assistant can create:

• Google Sheets tables
• Google Docs
• CSV files
• Presentations

How do i create a quarterly report?

Use a step-by-step approach:

1. Collect data:

Show view statistics for Q4 2025

2. Analyze:

What were the 10 most popular videos?
Which videos have low completion rates?

3. Create a document:

Create a Google Presentation with this data and conclusions

Detailed examples: Usage Examples

Limitations and performance

Are there limits on the number of requests?

Yes, standard Kinescope API rate limits apply:

• Business plan: up to 100 requests per minute
• Enterprise plan: up to 500 requests per minute

When the limit is exceeded, you will get an error suggesting to retry after a few seconds.

Recommendations:

• Formulate queries precisely to get the needed result on the first try
• For bulk operations, pause between queries
• Consider upgrading to a higher plan

Why is MCP working slowly?

Possible causes:

1. Large data volume — query processes many videos
2. Complex analytics — metric calculations take time
3. Server load — high load at the moment
4. Rate limiting — request limit reached

Solution:

• Clarify the query, add filters
• Break into several smaller queries
• Wait a few seconds
• Use a higher-tier plan

Can i use MCP for bulk operations?

Yes, but with limitations:

What is possible:

• Updating metadata for multiple videos
• Bulk tag addition
• Moving videos between projects
• Creating reports on a large number of videos

Limitations:

• Observe rate limits
• For operations with thousands of videos, use the direct API
• Delete operations are unavailable

What Kinescope plan is needed for MCP?

Minimum: any Kinescope plan

Recommended: Business or higher

Why Business or higher:

• Higher rate limits (more requests)
• Access to all analytics metrics
• Advanced features (DRM, SSO, etc.)
• Priority support

Security and access rights

Who can use my MCP connector?

Only you, if you have not shared the connection settings.

For team collaboration:

• Each team member configures their own connector
• Access rights are determined in Kinescope
• Members see only their own projects

What can AI do with my videos?

AI can only do what is permitted by your rights in Kinescope:

• View information
• Edit metadata
• Create projects and playlists
• Get analytics
• Cannot delete data (disabled in MCP)
• Cannot change account settings
• Cannot manage payments

Are actions logged through MCP?

Yes, all actions through MCP Server are recorded in your Kinescope account logs, just like actions through the web interface.

What is logged:

• information viewing
• Metadata changes
• Structure creation and modification
• Request time and IP address

Can i revoke access?

Yes, simply delete or disable the MCP server in your AI assmstant’s settings (Cursor or Qwen3). Access through MCP will be immediately terminated.

integrations

Can i integrate with Google Drive?

Yes, AI assistants can work simultaneously with Kinescope MCP and other services:

Get video data from Kinescope and create a report in Google Docs

Upload a list of videos from Google Sheets and update their metadata in Kinescope

Does MCP work with other platforms?

MCP Server allows the AI assistant to work with multiple services simultaneously.

integration examples:

• Google Workspace (Docs, Sheets, Drive)
• Slack (notifications and reports)
• Notion (documentation)
• And other services supported by your AI assistant

Can i automate tasks?

Partially. AI assistants with MCP support are great for:

• Creating regular reports
• On-demand data analysis
• Bulk metadata updates

For full automation (scheduled tasks, webhooks), use Kinescope API .

Troubleshooting

Getting error 401 (Unauthorized)

Cause: authorization issues

Solution:

1. Check that you are authorized in Kinescope
2. Reconnect the connector in Claude
3. Check the Kinescope account status
4. Contact the support chat within the Kinescope interface

Getting error 429 (Too Many Requests)

Cause: rate limit exceeded

Solution:

1. Wait 30-60 seconds
2. Reduce the request frequency
3. Break the large task into several smaller ones
4. Consider upgrading to a higher plan

AI does not understand my query

Cause: the query is worded ambiguously

Solution:

1. Be more specific: specify iDs, tags, dates
2. Break a complex query into simple steps
3. Use examples from documentation
4. Try rephrasing

Data is not updating

Cause: caching or processing delay

Solution:

1. Wait 1-2 minutes
2. Make the query again
3. Check data in the web interface
4. Contact the support chat within the Kinescope interface if the issue persists

What’s next?

Additional resources

1. Kinescope MCP Server — complete guide
2. Usage Examples — ready-made scenarios
3. API Documentation — technical documentation
4. MCP Protocol — official protocol documentation

Dmdn’t find an answer to your question?

• the support chat within the Kinescope interface — quick answers from specialists
• Email — for detailed questions
• GmtHub — technical documentation

Still have questions? Write to the support chat within the Kinescope interface — our specialists will help!