Agora MCP for AI. Control streaming and recordings with your agent.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Connect to your AI in seconds.
Agora MCP manages real-time video and voice communications directly from your agent. It gives you programmatic control over streaming sessions, allowing you to list active users in channels, start or stop cloud recordings, and monitor granular usage data like bandwidth consumption.
You can also manage the underlying projects and configure whiteboard rooms without ever leaving your development environment.
What your AI can do
Acquire cloud recording
Requests a resource ID needed before you can begin cloud recording for an Agora session.
Create project
Initializes and creates entirely new Agora projects within the platform.
Get project details
Retrieves specific, detailed information about a chosen project's configuration.
Create and retrieve detailed configurations for Agora projects, ensuring your agent has the correct credentials.
List all users currently connected to a specific channel or whiteboarding room.
Initiate, check the status of, and terminate cloud recording sessions for auditing purposes.
Retrieve usage data detailing consumed minutes, bandwidth peaks, and concurrent user counts over time.
Set up channels and adjust specific rules to optimize the real-time interactive experience.
Ask an AI about this
Waiting for input…
Agora: 10 Tools for Stream Management
Use these tools to automate everything from project setup and channel monitoring to starting and querying cloud recording status.
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 Agora on VinkiusAcquire Cloud Recording
Requests a resource ID needed before you can begin cloud recording for an Agora session.
Create Project
Initializes and creates entirely new Agora projects within the platform.
Get Project Details
Retrieves specific, detailed information about a chosen project's configuration.
Get Usage Stats
Queries and returns usage data for a specific project, detailing consumed resources.
List Active Calls
Retrieves a list of all ongoing real-time communication calls and active sessions.
List Projects
Lists every Agora project that has been set up in your console for quick reference.
List Whiteboard Rooms
Shows a list of all active interactive whiteboard rooms currently running.
Query Cloud Recording
Checks the current status and progress of an ongoing cloud recording task.
Start Cloud Recording
Begins recording a specified channel, sending the stream data to the cloud for later...
Stop Cloud Recording
Immediately halts an active cloud recording session that is currently running.
Security and governance baked right in.
Pick your AI client below to get set up. Just create a Vinkius account, subscribe, and you're instantly up and running. We handle the entire backend infrastructure, delivering out-of-the-box support for HTTPS Streamable, SSE, and OAuth2—zero messy routing required.
Claude AI
Open Claude Settings
Go to claude.ai, click your profile icon, then navigate to Customize → Connectors.
Add Custom Connector
Click the "+" button and select Add custom connector. Paste your Vinkius endpoint URL:
https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp
Replace [YOUR_TOKEN_HERE] with your token
from cloud.vinkius.com. For OAuth-protected servers, expand
Advanced settings to add credentials.
Start a conversation
Open a new chat. The Agora integration is available immediately — no restart needed.
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 Agora, then connect any of our 5,100+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,100+ 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 Agora. 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 connection provides 10 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.
Getting a full picture of real-time usage is always a nightmare.
Today, figuring out how much bandwidth or compute power your streaming service used feels like detective work. You have to log into the console, click through multiple tabs, select date ranges, and then copy numbers from different graphs just to get a total usage report for billing or capacity planning.
With this MCP, you skip all that clicking. Your agent runs `get_usage_stats` directly. You ask, 'What was our peak concurrency last quarter?' and it hands you the number—no dashboards, no copy-pasting required.
Start your recording workflow with the start_cloud_recording tool.
Before this MCP, starting a recorded session meant multiple manual steps: first generating temporary credentials, then initiating the stream in one panel, and finally tracking it across another monitoring dashboard. It was always a multi-page, high-friction process.
Now, your agent coordinates the entire sequence for you. You tell it to record 'webinar_live,' and it handles the necessary calls—like `acquire_cloud_recording` first—then executes `start_cloud_recording`. The whole workflow runs in a single conversational turn.
What your AI can actually do with this
This MCP connects your AI client to Agora's platform for video, voice, and interactive streaming services. Instead of logging into a console or writing complex SDK calls, you just ask your agent what needs to happen. You can programmatically create and manage communication channels; need to archive a meeting? Your agent handles the cloud recording lifecycle—starting it, monitoring its status, and stopping it when done.
It also pulls detailed project metrics on usage and bandwidth consumption. If you're managing multiple systems, Vinkius makes sure that connection remains stable, giving your agent access to all these core functions in one place. This means you can automate everything from checking channel occupancy to generating compliance reports based on recorded sessions.
019d8413-bc01-7252-820b-815089738430 
Here's how it actually works
The bottom line is that you tell your AI client what state you need—whether it's a report or an active recording—and it runs the necessary sequence of calls to make it happen.
First, your agent uses list_projects to identify all available Agora projects in your console.
Next, you direct your agent to use a function like acquire_cloud_recording to secure the necessary resource ID for recording.
Finally, your agent executes start_cloud_recording with the required credentials and channel details, immediately giving you confirmation of the active session.
Who is this actually for?
This MCP serves Ops Engineers, Full-Stack Developers, and Product Managers who get frustrated having to jump between multiple cloud consoles just to check usage or manage sessions. It gives you the ability to automate complex workflows without manual intervention.
Automates channel provisioning; they use this MCP to provision communication channels and record the setup process for new client environments.
Needs a single place to monitor system health across multiple projects, checking peak usage metrics via get_usage_stats when issues arise at 2 AM.
Designs and tests the full lifecycle of streaming features; they use it to manage project credentials (create_project) before building any client-facing feature.
What Changes When You Connect
You get immediate visibility into usage. Instead of digging through dashboards, running get_usage_stats gives you the exact video minutes and bandwidth consumed for a given project.
Archival is simple. You can start recording a session using start_cloud_recording, and later confirm its status with query_cloud_recording, all via natural language prompts to your agent.
Channel management is instant. If you need to know who's in the meeting, running list_active_calls gives you a real-time roster of UIDs without writing any boilerplate code.
Setup and maintenance are streamlined. Use create_project when starting a new client integration; your agent handles the initial project setup flow for you.
Debugging is faster. Need to know what rooms are active? Running list_whiteboard_rooms instantly lists every interactive session running right now.
See it in action
Compliance Audit After a Training Session
The PM needs to prove that all 20 participants were present and recorded the discussion. They ask their agent to start_cloud_recording for the specific channel, wait until the session ends, then use stop_cloud_recording before querying the final record status with query_cloud_recording.
Capacity Planning Review
The Ops Engineer needs to predict next month's bill. They tell their agent to run get_usage_stats for the last quarter, which pulls detailed metrics on minutes consumed and peak concurrency, allowing immediate budget adjustments.
Real-Time Troubleshooting
A developer sees an intermittent connectivity issue. They ask their agent to run list_active_calls immediately to see if the failure correlates with a sudden spike in concurrent users or active whiteboard rooms, helping pinpoint the exact moment of failure.
Initial Client Integration
A full-stack developer is setting up a new client. They use their agent to create_project first, then ask it to pull all necessary credentials using get_project_details so they can plug them into the application code.
The honest tradeoffs
Treating API calls as disconnected steps
The user runs a script that just calls list_active_calls, gets a list, and then has to manually write separate logic blocks for each specific action like starting recording or checking usage.
Don't chain functions with basic scripts. Instead, let your agent orchestrate the entire flow: First, check if recording is needed; if yes, run acquire_cloud_recording and then use start_cloud_recording. This makes the entire process atomic.
Over-relying on console UI
A manager has to click through Project Management, select the app ID, then navigate to Usage Reports, only to copy a single number of bandwidth usage.
The agent uses get_usage_stats directly. You ask for 'last month's peak bandwidth,' and it returns the data point instantly, eliminating all manual clicks.
Confusing listing with retrieving
Trying to figure out project details by running list_projects and then having to manually look up the specific AppID from a massive list.
Always use get_project_details after using list_projects. This pulls all relevant, structured data for one project into your agent's context immediately.
When It Fits, When It Doesn't
Use this MCP if you need to manage the full lifecycle of real-time communication. That means needing to list users in a channel (list_active_calls), monitor resource usage over time (get_usage_stats), or archive content using cloud recording tools (start_cloud_recording and stop_cloud_recording). It's ideal for compliance teams, ops engineers, and anyone building streaming features. Don’t use it if your only goal is simple messaging—that requires a different type of connector. If you just need to know the current status of one resource without calling any functions, stick to reading documentation. But if you must act on the data, like provisioning credentials (create_project), this MCP has the tools for that.
Questions you might have
How do I check who is on an active channel using list_active_calls? +
You simply ask your agent to run list_active_calls for the specific room. It pulls a real-time roster, giving you UIDs and connection status without needing manual UI interaction.
What is the difference between list_projects and get_project_details? +
list_projects gives you a master list of every project ID. You then use get_project_details to pull all the specific, required data—like AppIDs or token settings—for one selected project.
Can I stop a recording using stop_cloud_recording? +
Yes. If a session is finished early, you can use stop_cloud_recording. This function immediately halts the stream and prevents unnecessary billing or data transfer.
Do I need to run acquire_cloud_recording before start_cloud_recording? +
Yes. You must first call acquire_cloud_recording to secure a unique resource ID. This ID is required input for the start_cloud_recording tool to function.
When I use `create_project`, what does the tool handle for my Agora environment? +
It provisions a completely isolated Agora project structure. This ensures that your testing and production environments remain separate, preventing accidental data overlap or configuration conflicts between different applications.
When I use `get_usage_stats`, what specific dimensions of usage can I track? +
The tool tracks detailed consumption across video minutes, audio minutes, and peak concurrent users. You get granular data points that help you pinpoint exactly where your bandwidth or compute costs are peaking within the project timeframe.
How do I check which interactive session is currently active using `list_whiteboard_rooms`? +
This tool provides a real-time list of all active Interactive Whiteboard sessions. You can quickly identify the unique IDs and connection statuses for collaborative rooms without having to manually monitor the platform console.
What critical information does `get_project_details` return about my project's setup? +
It returns comprehensive details, including the AppIDs and token settings for your specified project. This is crucial for ensuring that your AI agent has the correct credentials and scope to interact with all necessary Agora services securely.
Can I automatically list all active users in an Agora channel? +
Yes! Use the list_channel_users tool with the project AppID and the target Channel Name. Your agent will return a list of UIDs currently connected to that session.
How do I start a cloud recording session via the AI agent? +
Use the start_cloud_recording tool with the necessary parameters (AppID, Channel Name, and recording config). The agent will trigger the Agora cloud recording engine and return the Resource ID and SID.
Can I monitor the total minutes consumed by my project this month? +
Yes! The get_usage_metrics tool retrieves granular consumption data, including audio and video minutes, filtered by date range or project.
We've already built the connector for Agora. Just plug in your AI agents and start using Vinkius.
No hosting. No infrastructure. No complex setup.
All 10 tools are live and waiting.
You're up and running in seconds.
Gemini Vinkius gives your AI agents access to the full catalog of app connectors, all fully managed, secure, and enterprise-ready. One subscription, every tool you need.
Built, hosted, and secured by Vinkius. You just connect and go.