Bland AI MCP. Automate Hyper-Realistic Voice Outreach.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Bland AI handles all your automated phone communication. It lets you deploy realistic, voice-driven agents that manage everything from outbound qualification calls to complex customer support interactions.
You build persistent AI personas with specific personalities and voices, then trigger them programmatically for any country. Every conversation is recorded and analyzed afterward, giving you immediate transcripts and actionable insights without manual effort.
What your AI agents can do
Analyze call transcript
Runs post-call analysis on a transcript to extract specific variables or sentiment summaries.
Create voice agent
Builds and deploys a new, persistent AI persona with fixed prompts and voice settings.
Delete voice agent
Removes an existing configured AI persona from the system.
Send programmatic phone calls to specific numbers using configured AI agents.
Create, update, and delete persistent AI profiles that maintain a consistent brand voice during conversations.
Get real-time details about ongoing calls or manually stop an active session.
Fetch full call transcripts and metadata for any completed interaction.
Run post-call jobs to extract structured variables, insights, or sentiment summaries from the transcript.
Ask AI about this MCP
Supported MCP Clients
OAuth 2.0 Compatible Gemini Waiting for input…
Bland AI: 12 Tools for Voice Automation
These tools let you programmatically create, deploy, monitor, and extract data from high-fidelity AI phone calls.
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 Bland AI on Vinkius019dd0c2analyze call transcript
Runs post-call analysis on a transcript to extract specific variables or sentiment summaries.
019dd0c2create voice agent
Builds and deploys a new, persistent AI persona with fixed prompts and voice settings.
019dd0c2delete voice agent
Removes an existing configured AI persona from the system.
019dd0c2get agent config
Retrieves the current settings and configuration details for a specific AI agent.
019dd0c2get call details
Fetches detailed information, including the full high-fidelity transcript, for a given call ID.
019dd0c2list available voices
Lists all available high-fidelity AI voices you can assign to an agent.
019dd0c2list phone numbers
Displays the directory of purchased phone numbers associated with your account.
019dd0c2list recent calls
Retrieves a list of recently completed or ongoing phone calls.
019dd0c2list voice agents
Shows all currently configured and managed AI personas (voices agents).
019dd0c2send phone call
Initiates an outbound phone call using a specified AI agent to a target number.
019dd0c2stop active call
Halts an ongoing, live phone conversation immediately.
019dd0c2update agent config
Modifies the operational settings or behavioral prompts of an existing AI agent.
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 Bland AI, 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 Bland AI. 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 12 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Reviewing Call Data Takes Too Long Today
Today, running a campaign means dialing dozens of numbers and then spending hours playing back recordings. You have to manually scrub the audio: 'Did they mention pricing? Did they sound hesitant? What was their title?' It's copy-paste hell across multiple tabs.
With this MCP, you simply send the call via `send_phone_call`. When it’s done, your agent handles everything. You get a clean transcript and can run post-processing right away with `analyze_call_transcript` to pull out every variable you need.
Getting Agent Configuration Details
Manually checking agent instructions requires logging into the platform and clicking through multiple settings pages, hoping nothing changed since last week.
Now, you just call `get_agent_config`. You get all the current rules, prompts, and personality settings in one clean data payload. It's immediate and reliable.
What you can do with this MCP connector
You stop relying on manual dialing or listening through hours of recordings just to find a single piece of data point. This MCP lets your AI act as a dedicated voice engineer, handling all inbound and outbound calls with natural conversation. You define the agent's personality, set fixed prompts, and manage it—all before making a call.
When you initiate an outreach sequence, the system executes high-fidelity calls to over 200 countries, following precise instructions given by your agent.
After the call ends, you don't just get a recording link; you get immediate access to the full transcript and can run post-call analysis to pull specific data points or sentiment summaries. If you build multi-step automations—say, calling a lead, getting the transcript, then having another agent automatically log that outcome into your CRM—you use the power of Vinkius AI Analytics.
This platform gives you full visibility into every single tool call and data flow, so nothing happens in the dark. It’s how you manage an entire voice communication ecosystem from one place.
019dd0c2-9d64-70a2-aa7a-6958d508f34a 
How Bland AI MCP Works
- 1 First, you use
list_available_voicesandlist_phone_numbersto check what voices and phone lines are ready. - 2 Next, you use
create_voice_agentto build your AI persona, giving it a fixed prompt and personality settings. You can then adjust its behavior usingupdate_agent_configor view its settings withget_agent_config. - 3 Finally, you trigger the outreach by calling
send_phone_call. When that's done, you uselist_recent_callsandget_call_detailsto pull transcripts for analysis.
The bottom line is: You define the agent behavior once, and then your agent handles complex voice interactions across any platform connected through Vinkius.
Who Is Bland AI MCP For?
This MCP is built for Operations Leads and Sales Directors who are tired of managing phone outreach via spreadsheets or manually reviewing call recordings. It’s for the developer needing reliable, high-volume voice interactions in their custom workflows.
They use this to automate initial lead qualification calls, confirming meeting availability and gathering basic buyer intent data.
They run automated appointment reminders or follow-up surveys with existing clients without having to pick up the phone themselves.
They integrate high-speed, conversational AI into complex business pipelines that need external voice touchpoints.
What Changes When You Connect
- Never manually listen for insights again. After a call, run
analyze_call_transcriptto extract variables or sentiment summaries instantly. - Maintain brand consistency by creating persistent personas with
create_voice_agent. The agent acts exactly like you define it every time. - Manage your entire calling infrastructure—from listing numbers (
list_phone_numbers) to selecting the voice (list_available_voices)—all through one interface. - Get immediate proof of performance. Use
get_call_detailsto pull transcripts and recordings right after a call finishes, without leaving the workflow. - Build complex automations by chaining this MCP with others. You can run calls, get data, and then send that structured result elsewhere using Vinkius's cross-MCP capabilities.
Real-World Use Cases
Lead Qualification Follow-Up
A sales rep needs to confirm if a prospect is still interested in the product. Instead of manual dialing, they call send_phone_call via their agent. The agent confirms interest and logs the outcome automatically using analyze_call_transcript.
Customer Churn Prevention
A CSM needs to check in with an at-risk client. They call send_phone_call, instructing the agent to diagnose potential issues. Afterward, they use get_call_details to review the full transcript and pinpoint the exact conversation turning point.
Mass Appointment Reminders
An operations team needs to remind 100 people about a meeting. They list their numbers with list_phone_numbers and run an automated loop, calling send_phone_call for each contact until the confirmation is received.
Debugging Agent Behavior
A developer needs to see why a call failed. They check the logs with list_recent_calls, then use get_call_details to review the transcript and pinpoint if the failure was due to bad data or poor agent instructions.
The Tradeoffs
Treating calls as simple scripts
Assuming that just because you send a call, the conversation is over. You end up manually reading recordings for hours to find out if they bought something or what their pain point was.
→
You must use analyze_call_transcript right after the call completes. This tool pulls structured data (like sentiment score or next steps) from the transcript, turning unstructured talk into usable fields.
Ignoring agent setup
Just sending a call without defining who is calling and why. The conversation will sound random, unprofessional, and fail to build trust.
→
Always start by using create_voice_agent to define the persona. This ensures that every interaction maintains a fixed brand voice and professional tone.
Over-relying on raw recordings
Downloading hours of audio files into a spreadsheet just to manually copy/paste key dates or names.
→
Use get_call_details first. This retrieves the high-fidelity transcript, which is the source material you pass to analyze_call_transcript for clean data extraction.
When It Fits, When It Doesn't
You should use this MCP if your primary bottleneck is converting unstructured voice conversations into structured business data. Specifically, if you need reliable, high-volume outreach (use send_phone_call) and then need to extract actionable metrics from the results (use analyze_call_transcript). Don't use it if your only goal is simple SMS messaging; for that, a dedicated chat MCP will be cleaner. If all you want is to view past calls without any analysis, just calling list_recent_calls gives you enough data. But if you need the insight from those calls, this is what you need.
Common Questions About Bland AI MCP
How do I find my Bland AI API Key? +
Log in to your account, navigate to Settings > API Keys, and generate a new key for your integration.
Can I use specific AI voices via agent? +
Yes! The send_phone_call and create_voice_agent tools allow you to specify voice IDs (e.g., 'maya', 'mason') programmatically.
How do I retrieve call transcripts? +
Use the get_call_details tool with a specific call ID to retrieve the high-fidelity transcript and recording link directly from the platform.
When I run `analyze_call_transcript`, what specific insights can I extract from a call? +
It pulls out structured data points like sentiment summaries, key variables, or conversion likelihood. You feed it the transcript and tell it exactly what kind of analysis you need.
If an agent gets off-script mid-call, how do I use `stop_active_call` to regain control? +
You can immediately halt any ongoing phone conversation using stop_active_call. This is useful if the AI deviates from the script or you need to manually intervene in the call flow.
How do I manage my configured personas after listing them with `list_voice_agents`? +
You can modify an existing agent's prompt using get_agent_config, or update its instructions via update_agent_config. If you need to remove it, use delete_voice_agent.
Can I use `send_phone_call` for large-scale outbound campaigns? +
Yes. The tool initiates high-fidelity calls to over 200 countries. This lets you orchestrate lead qualification or appointment setting across a massive number of contacts.
How do I check my available resources before running an outreach campaign? +
Run list_phone_numbers to verify your purchased numbers, and use list_available_voices to confirm which high-fidelity AI voices are ready for deployment.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.