MCP SERVER · OFFICIAL PROJECTSearch and analyse a Kaltura media library read-only — entries and captions, download and thumbnail URLs, retention curves, real-time and geographic analytics.
LABSCO SUMMARY
get_video_retention shows where viewers actually leave a video, which is a different question from how many watched it, and caption content turns a media library into something searchable by what was said rather than by title. The analytics tools are split by purpose on purpose — time series for charts, real-time for monitoring, quality metrics for streaming problems — so you get the shape you need instead of a generic table to reshape. Read-only throughout: it will not change or delete anything in your account, which makes it safe to attach to a production Kaltura partner.
A read-only MCP server for the Kaltura video platform. It covers discovery (searching entries, browsing categories, reading captions and attachments) and a full analytics suite built as purpose-specific tools rather than one generic report call. Nothing it exposes writes back to Kaltura.
search_entries, get_media_entrykaltura://categories/tree — list_categorieslist_caption_assets, get_caption_contentlist_attachment_assets, get_attachment_contentget_download_url, get_thumbnail_urlget_analytics, get_analytics_timeseries, get_video_retention, get_realtime_metrics, get_quality_metrics, get_geographic_breakdownkaltura://analytics/capabilities — list_analytics_capabilitieskaltura://media/recent/{count}, up to 100 entries, cached for 5 minutesA Kaltura account and its API credentials from KMC → Settings → Integration Settings: the service URL, your numeric partner ID, the admin secret and a user ID. Locally these go in a .env file the server loads from the project directory — run setup_env.py to write it with 600 permissions and print the exact client configuration. The project is kaltura-mcp (version 0.2.0), a name PyPI does not have, and it provides kaltura-mcp for stdio, kaltura-mcp-remote for the hosted HTTP/SSE server and kaltura-mcp-proxy to bridge a remote server into a stdio client. Hosting it remotely additionally needs JWT_SECRET_KEY, OAUTH_REDIRECT_URI, SERVER_HOST and SERVER_PORT; users then authorize through a web form and receive a JWT that expires after 24 hours, with credentials encrypted inside it.
Build from source — git clone https://github.com/zoharbabin/kaltura-mcp.git
USE CASES
Search entries, list the captions on the candidates, and read the caption content to confirm before opening anything.
Pull the retention curve for the video and read the drop-off points, rather than inferring engagement from a view count.
Watch real-time metrics alongside quality-of-experience data, so a drop in viewers can be separated from a streaming problem.
Break viewing down by country, region or city and pull the matching time series for a chart.
Use the accessibility prompt to walk the entries and their caption assets, and get back which videos have no subtitles.
Fetch the direct download URL for the file and a thumbnail at the dimensions and timestamp you need.
Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.
get_analyticsGet detailed analytics in TABLE format for reporting. USE WHEN: Creating reports, comparing metrics, ranking content, analyzing performance, exporting data. RETURNS: Structured data with headers/rows. EXAMPLES: 'Show top 10 videos by views', 'Compare user engagement by category', 'Export monthly performance report'. Use list_analytics_capabilities to see all 60+ report types. For charts/graphs, use get_analytics_timeseries instead.
get_analytics_timeseriesGet time-series analytics data optimized for charts and visualizations. Use this when creating graphs, dashboards, or tracking trends over time.
get_attachment_contentDownload or read ATTACHED FILES from videos. USE WHEN: Accessing supplementary materials, downloading PDFs, getting presentation slides, reading attached documents. RETURNS: File content (if text) or download URL. EXAMPLE: 'Download the PDF slides', 'Read the attached notes'. Use after list_attachment_assets to get specific attachment ID.
get_caption_contentGet actual CAPTION TEXT or download captions file. USE WHEN: Reading video transcript, downloading subtitles, analyzing spoken content, creating accessible content. RETURNS: Full caption text and download URL. EXAMPLE: 'Get English subtitles for video', 'Read transcript to find mentions of topic'. Use after list_caption_assets to get specific caption ID.
get_download_urlGet direct DOWNLOAD link for video files. USE WHEN: User needs to download/save video locally, export for editing, backup content, share downloadable link. RETURNS: Time-limited secure URL for downloading. EXAMPLE: 'Download video 1_abc123', 'Get mp4 file for editing'. Different from streaming - this is for saving files.
get_geographic_breakdownAnalyze viewer LOCATIONS and regional performance. USE WHEN: Understanding global reach, planning regional strategies, checking market penetration, optimizing CDN, compliance checks. RETURNS: Views/viewers by country/region/city with percentages. EXAMPLES: 'Which countries watch our content?', 'Show US state breakdown', 'Find top 10 cities for viewership'. Includes map-ready data.
get_media_entryGet complete metadata for a single video/media file. USE WHEN: You have a specific entry_id and need full details (title, description, duration, tags, thumbnail, status). RETURNS: Complete media metadata including URLs, dimensions, creation date. EXAMPLE: After search finds entry_id='1_abc123', use this to get full video details.
get_quality_metricsAnalyze streaming QUALITY and viewer experience. USE WHEN: Troubleshooting playback issues, monitoring streaming performance, optimizing delivery, investigating viewer complaints. RETURNS: Buffer rates, bitrate averages, error rates, startup times, quality scores. EXAMPLES: 'Why are users complaining about buffering?', 'Check streaming quality by device type', 'Find videos with poor performance'. Helps ensure smooth playback.
get_realtime_metricsGet LIVE analytics updating every 30 seconds. USE WHEN: Monitoring live events/streams, building real-time dashboards, tracking immediate campaign impact, detecting issues as they happen. RETURNS: Current active viewers, plays per minute, bandwidth usage. EXAMPLES: 'How many people watching right now?', 'Monitor live event performance', 'Track viral video in real-time'. Different from historical analytics - this is NOW.
get_thumbnail_urlGet video THUMBNAIL/POSTER image. USE WHEN: Displaying video previews, creating galleries, showing video cards, generating custom thumbnails. RETURNS: Image URL with your specified size. EXAMPLES: 'Get thumbnail for video 1_abc123', 'Create 400x300 preview image', 'Get frame from 30 seconds in'. Can capture any frame from video.
get_video_retentionAnalyze WHERE viewers stop watching in a video. USE WHEN: Optimizing video content, finding boring sections, identifying engaging moments, improving completion rates. RETURNS: 101 data points (0-100%) showing viewer count at each percent of video. EXAMPLES: 'Where do viewers drop off in video 1_abc123?', 'What parts get replayed?', 'Compare retention for anonymous vs logged-in users'. Shows exact percentages where audience is lost.
list_analytics_capabilitiesDiscover ALL analytics capabilities of this system. USE WHEN: User asks 'what analytics can you do?', exploring available reports, understanding metrics options, learning about analytics features. RETURNS: Complete list of 7 analytics functions with descriptions, 60+ report types, available dimensions, time intervals. EXAMPLE: Always run this when user first asks about analytics. No parameters needed - just call it!
list_attachment_assetsFind FILES ATTACHED to videos. USE WHEN: Looking for supplementary materials, PDFs, slides, documents linked to video. RETURNS: List of attached files with names, types, sizes, IDs. EXAMPLES: 'What documents are attached to training video?', 'Find PDF slides for presentation'. Attachments are additional files uploaded with videos.
list_caption_assetsFind all CAPTIONS/SUBTITLES for a video. USE WHEN: Checking if video has captions, finding available languages, preparing for accessibility, getting transcript. RETURNS: List of caption files with languages, formats (SRT/VTT), IDs. EXAMPLE: 'Does video 1_abc123 have captions?', 'List subtitle languages available'. First step before getting caption content.
list_categoriesBrowse content organization hierarchy. USE WHEN: Exploring content structure, finding category IDs for filtering, understanding content taxonomy. Categories organize videos into folders/topics. RETURNS: Tree structure with category names, IDs, parent-child relationships. EXAMPLE: Find all videos in 'Training' category by first getting category ID.
search_entriesSEARCH for videos or LIST all content. USE WHEN: Finding videos by keyword, listing newest content, discovering what's available, filtering by date/category. POWERFUL SEARCH across titles, descriptions, tags, captions. EXAMPLES: 'Find videos about python', 'Show newest 10 videos' (use query='*' sort_field='created_at'), 'Search in transcripts for keyword', 'List videos from last week'. This is your primary discovery tool!
BEFORE YOU INSTALL
16 tools when we connected.
The documented client runs the server on your machine.
CONTEXT COST
Rated by the size of the 16 tool definitions that ride along on every turn, not by how many there are
INSTALLATION
You clone this one yourself.
git clone https://github.com/zoharbabin/kaltura-mcp.git
cd kaltura-mcp
pip install -e .
python setup_env.py
which kaltura-mcp{
"mcpServers": {
"kaltura": {
"command": "<YOUR_PATH_TO_KALTURA_MCP>"
}
}
}MCP servers
Kush36Agrawal
samuelgursky
Dakkshin
burningion