Loom MCP. Control async videos and viewer metrics from chat.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Loom (Async Video Messaging) MCP Server lets your AI agent manage all aspects of video communication. Retrieve detailed metadata, track viewer analytics for shared content, and read or add threaded comments on any video timeline—all through natural conversation.
It also handles generating raw MP4 download links and updating video properties without opening the web UI.
What your AI agents can do
Delete video
Permanently deletes a specific screencast video from your Loom workspace. This action cannot be undone.
Get transcript
Retrieves the full text transcript for any given video file.
Get video
Fetches basic metadata, including title and duration, for a specific Loom video.
Get detailed viewer data—watch segments and total viewership counts—using the video analytics tool.
Read, add, or remove threaded comments directly on a video timeline without opening Loom.
Fetch titles, durations, permalinks, and transcripts for all videos in your workspace using list_videos or get_video.
Generate a precise, temporary download link for the raw MP4 file of any video.
Update titles or move specific screencasts into target folders using update_video and list_folders.
Ask AI about this MCP
Supported MCP Clients
OAuth 2.0 Compatible Gemini Waiting for input…
Loom (Async Video Messaging) MCP Server: 8 Tools for Video Ops
These tools allow your AI agent to programmatically interact with Loom. You can manage the entire video lifecycle, from listing assets and getting transcripts to updating properties and deleting content.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using Loom (Async Video Messaging) on Vinkius019d75cadelete video
Permanently deletes a specific screencast video from your Loom workspace. This action cannot be undone.
019d75caget transcript
Retrieves the full text transcript for any given video file.
019d75caget video
Fetches basic metadata, including title and duration, for a specific Loom video.
019d75caget video analytics
Retrieves detailed usage data, showing exactly which segments of the video were watched by viewers.
019d75calist folders
Lists all existing organized folders within your Loom workspace for reference and organization.
019d75calist videos
Returns a list of every video currently accessible in the specified Loom workspace.
019d75calist workspace members
Lists all users who are members of the connected Loom workspace.
019d75caupdate video
Changes the title or properties of an existing video asset in your library.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Loom (Async Video Messaging), then connect any of our 4,800+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,800+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Loom. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 8 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Managing a library of screencasts means endless context switching.
Today's process for auditing shared content is brutal. You open Loom, navigate to the video list, click into Video A to copy its permalink; then you switch tabs to check comments in a separate window. To get analytics, you have to hit another tab and wait for the metrics page to load—all while tracking what you copied and where.
With this MCP server, you just ask your agent: 'Give me the list of videos and their watch segments.' The system handles the multi-step process, pulling metadata, running `get_video_analytics`, and consolidating everything into a single response. It's instant.
Loom (Async Video Messaging) MCP Server
Before, updating a video title or moving it to an archive folder meant navigating through the UI and manually clicking 'Save'. If you needed to delete old content, it was a multi-step process of finding the asset and hitting the trash can icon.
Now, tell your agent: 'Update the Q1 pitch title to include the client name.' The server executes `update_video` directly. You treat your video library like data—manageable through simple commands.
What you can do with this MCP connector
Your agent handles every damn thing with Loom, so you don't gotta mess around in the web UI. You can manage your entire video workflow—from checking who watched what to cleaning up old files and getting raw media links—all through talking to your AI client.
You need an inventory? You can run list_videos to get a full rundown of every single screencast you've got in the workspace. For more detail on any specific clip, get_video fetches basic metadata right away, giving you things like the title, the total duration, and the permalink for that video.
For organizing your library? You can check out what folders exist by running list_folders. If a video needs sorting or renaming, you use update_video; this lets you change the asset's properties or update its title directly. And if things get messy, there’s delete_video, which permanently wipes that screencast from your workspace—it ain't reversible.
To really dig into engagement, you've got two key tools. First, run get_video_analytics to pull detailed usage data. This tells you exactly which segments of the video viewers watched and gives you the total viewership counts for that content. Second, if you need the full text breakdown, get_transcript retrieves the complete written transcript for any given file.
Collaboration is simple. You don't gotta open a browser tab to talk about the videos; your agent handles reading, adding, or removing threaded comments directly on the video timeline. For knowing who’s using the workspace, you can call list_workspace_members and get a list of every user connected.
Beyond basic management, you've got power tools for delivery. You can generate a precise, temporary download link for the raw MP4 file of any video asset. These capabilities let your agent take complete control, letting you update titles and properties using update_video while simultaneously pulling analytics or transcripts without ever leaving your chat window.
019d75ca-2b00-715e-a007-3a797539c37b 
How Loom MCP Works
- 1 Subscribe to this server and provide your Loom API Token.
- 2 Your AI client sends a natural language request (e.g., 'What did Sarah watch in the Q3 pitch video?').
- 3 The agent executes the necessary tools (
get_video,get_video_analytics) and returns the structured data to you.
The bottom line is, your AI client talks directly to Loom's core APIs so you don't have to use the web interface.
Who Is Loom MCP For?
Product managers who need to audit feature walkthroughs. Support reps dealing with high-volume customer video queries. Remote team leads monitoring project communication flow across dozens of shared videos. If you spend time tracking down what people watched or organizing the library, this is for you.
Audits viewer session data using get_video_analytics to determine which parts of a feature walkthrough need better documentation. Organizes feedback comments across project videos.
Responds to customer questions about shared video content by reading timeline comments and pulling up detailed metadata without switching screens.
Generates transcripts (get_transcript) of videos for documentation or uses update_video to rename assets based on new project specs.
What Changes When You Connect
- Stop guessing what people watched.
get_video_analyticsgives you precise segment telemetry, letting you know exactly where viewers dropped off or spent the most time. - Keep communication flowing without opening a browser. You can read or add threaded comments using your agent, keeping high-touch collaboration right in your chat window.
- Handle large libraries efficiently. Use
list_videosto get an inventory of every asset, and then useget_videofor the details you need on any specific one. - Access raw content instantly. The server generates ephemeral MP4 download endpoints, letting you grab the physical video file without fighting UI restrictions.
-
update_videolets you maintain library hygiene by renaming or relocating videos in real time, keeping your asset structure clean.
Real-World Use Cases
Auditing a Feature Demo
A PM needs to know if the new onboarding flow demo actually hit the 'Payment Setup' screen. They ask their agent to run get_video_analytics on the video ID, which returns session telemetry showing exactly how many viewers watched through that critical segment.
Handling a Support Query
A support rep gets asked about comments on an old training video. They prompt their agent to read the timeline using comment tools, instantly surfacing the conversation history without having to navigate deep into Loom's UI threads.
Cleaning Up Assets
The team lead realizes a series of outdated mockups are cluttering the main workspace. They prompt the agent to list_videos and then execute delete_video on the specific IDs, keeping the library optimized.
Archiving Content
A video needs a title change because the client rebranded. The content owner asks their agent to run update_video, changing the asset's name and then moving it into the 'Archive/2024' folder using the API.
The Tradeoffs
Copy-pasting metadata.
Logging into Loom, manually copying titles, durations, and permalinks from 50 different videos into a spreadsheet. This is tedious, error-prone work.
→
Use list_videos to pull the entire list of assets into your agent in one go. Then, use get_video for specific details you need, keeping everything automated.
Relying on UI downloads.
Needing a raw MP4 file for an external presentation but the Loom interface only provides limited preview versions or requires multiple clicks to download.
→ Ask your agent to generate the ephemeral MP4 download endpoint using its dedicated tool. You get the direct, raw link you need.
Missing context on engagement.
Knowing a video exists but not knowing if viewers actually paid attention or just skimmed the first 30 seconds.
→
Run get_video_analytics. This tool gives you the depth data, showing watch segments and actual viewership numbers, which is way better than just 'view count'.
When It Fits, When It Doesn't
Use this server if your workflow revolves around managing video assets, tracking viewer engagement, or maintaining collaborative comments within Loom. Specifically: If you need to know what people watched (analytics), who talked about it (comments/members), or where the files are (list videos/folders), you need these tools. Don't use this if your primary goal is simple file storage—a cloud bucket works better. Also, don't use it just to view content; you need an agent to perform actions like updating titles or deleting assets. The key differentiator is that the tool set allows programmatic control over the entire video lifecycle, from metadata retrieval (get_video) to ultimate deletion (delete_video).
Common Questions About Loom MCP
How do I get detailed metrics using get_video_analytics? +
You provide the specific Video ID to the agent, and it runs get_video_analytics. This returns session telemetry detailing exactly which watch segments were viewed by users.
Can I delete a video with delete_video? +
Yes. The tool uses delete_video to permanently remove the screencast from your workspace. Remember, this action is irreversible.
What can list_videos give me? +
list_videos returns a comprehensive inventory of every video in your workspace. You'll get enough data to start building an asset map or checking for missing content.
How do I make the raw MP4 file? Is there a download link? +
You ask the agent to generate an ephemeral download endpoint. It uses its internal mechanism to create a temporary, direct URL you can use to grab the raw MP4.
Can I read comments on a video using get_video_analytics? +
No. get_video_analytics only handles viewership data (segments watched). To see comments, you need to interact with the comment management tools directly.
How do I change a video's title or move it using update_video? +
You use update_video to rename content. This tool allows you to modify the video's visible title and properties in real-time without affecting the file itself. You can also relocate specific screencasts into different workspace folders for better organization.
What information does list_workspace_members provide? +
It lists every member associated with your Loom workspace. This helps you audit team roles and understand who has access to the content. It returns user profiles, which is useful for managing permissions or knowing key stakeholders.
How do I retrieve a full text copy using get_transcript? +
The get_transcript tool retrieves the complete spoken word transcript from a video file. It gives you clean, textual data that lets you index or search the content without needing to watch the recording.
Can I see who watched my video using my agent? +
Yes. Use the list_viewer_sessions tool with a specific Video ID. Your agent will retrieve the viewing telemetry, allowing you to see exactly how many times the video was watched and which specific segments were most engaging.
How do I download a raw MP4 file of a Loom video through a conversation? +
The create_download_url tool forces the backend to allocate an ephemeral S3 pre-signed URL. Your agent will provide this link, which allows you to stream or download the raw bytes of your video directly.
Can my agent respond to comments on my video timeline? +
Absolutely. Use the add_timeline_comment tool to post threaded messages directly to the video. Your agent can help you maintain community interaction by drafting and sending replies through natural conversation.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.