Thoughtly MCP. Run AI calls and manage CRM contacts from text.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Thoughtly connects your AI agent directly to a voice platform, giving it the ability to make calls and manage contacts in real-time.
Your agent can query CRM records, list agents, trigger outbound calls, and pull full call transcripts—all through simple text commands.
It bridges the gap between planning contact outreach and actually executing conversational phone workflows.
What your AI agents can do
Create contact
Adds a new contact record to the Thoughtly CRM, accepting names, emails, or phone numbers.
Delete contact
Removes an existing contact from your Thoughtsly directory.
Get call history
Retrieves the detailed transcript and metadata for a single, specific phone call ID.
Triggers an AI outbound phone call using a specified contact and agent persona.
Creates, reads, updates, or deletes specific records within the Thoughtly CRM directory.
Fetches detailed call logs and full conversation transcripts based on a Call ID.
Lists all active AI Voice Agents, allowing your agent to select the correct persona for a job.
Ask AI about this MCP
Supported MCP Clients
OAuth 2.0 Compatible Gemini Waiting for input…
Thoughtly: 11 Tools for Call and Contact Management
Manage every aspect of your communication workflow—from listing contacts to triggering live calls—with these specialized tools.
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 Thoughtly on Vinkius019dd173create contact
Adds a new contact record to the Thoughtly CRM, accepting names, emails, or phone numbers.
019dd173delete contact
Removes an existing contact from your Thoughtsly directory.
019dd173get call history
Retrieves the detailed transcript and metadata for a single, specific phone call ID.
019dd173get contact
Fetches all details associated with one known Thoughtly contact ID.
019dd173get interview
Retrieves the full profile and status of a specific AI Voice Agent persona.
019dd173list calls
Lists summary data for multiple past phone calls, acting as a call log overview.
019dd173list contacts
Retrieves a list of all contacts stored in the Thoughtly CRM.
019dd173list interviews
Lists all available AI Voice Agents (Interviews), showing their names and IDs.
019dd173trigger call
Dials a specified contact number using the selected AI voice agent persona for an outbound call.
019dd173update contact
Modifies existing details—like notes or status—for a specific Thoughtly contact record.
019dd173get user
Fetches basic user details for your connected Thoughtly account.
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 Thoughtly, then connect any of our 5,000+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,000+ 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 Thoughtly. 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 11 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Manually tracking follow-up conversations wastes hours every week.
Right now, after a call, you're stuck in the cycle. You have to open the phone system dashboard, find the Call ID, read through the transcript, copy key names and dates into your CRM, and then manually update the 'Next Action' field. It’s clicking across five different applications just to write one summary sentence.
With Thoughtly MCP Server, you give your agent access to `get_call_history`. The agent pulls the full conversation record, summarizes the outcomes (e.g., 'Booked 2 PM Tuesday'), and sends that structured data back. You get actionable context immediately, without ever opening a dashboard.
Thoughtly MCP Server: Control Contacts & Call History
You used to run through multiple steps just to qualify one lead: 1. Find the contact in the CRM. 2. Check if they're active. 3. Manually write a script and dial them from the phone app. This was slow, error-prone, and required three different logins.
Now, your agent handles it all. It first runs `list_contacts` to confirm existence. Then, it selects an agent persona via `get_interview`, and finally executes `trigger_call`. The entire sequence is managed by the prompt, allowing you to scale outbound campaigns instantly.
What you can do with this MCP connector
Thoughtly MCP Server: Your Agent on the Phone Line
Listen up. This server hooks your AI agent straight into Thoughtly’s voice platform. It lets you ditch text-only commands and turn simple inputs into real, two-way phone conversations. You're not just planning outreach; you're executing it. Your agent can talk to contacts, check out records, and pull full transcripts—all through plain language prompts.
Initiating Outbound Calls and Agent Personas
You gotta know who your agent is before they make a call. You start by checking the available personas using list_interviews, which spits out all the AI Voice Agents you've set up, giving you their names and IDs. Once you pick the right one, you can run through get_interview to pull that specific agent's full profile and status details—you gotta make sure they're ready to roll.
When it’s time to call someone, you use trigger_call. You just give your agent a contact number and the selected persona ID, and boom. It dials out. This tool runs the whole outbound conversation, making it feel like a natural chat while your AI client handles the back-end work.
Managing the CRM Directory
The server keeps track of every single person you talk to. You can treat the Thoughtly CRM like a living database right from your agent's prompt. Need a new lead? Use create_contact; just feed it names, emails, or phone numbers and that record is locked in.
Got a contact ID but need all their details—notes, status, everything? Run get_contact. If you know what’s wrong with an existing record, you can use update_contact to modify specific fields like notes or change the account's status. And if that lead is dead weight and you never want to see them again, you hit them with delete_contact.
For a quick overview of who you're dealing with, run list_contacts; it gives you the full rundown of every contact ID in the directory.
Reviewing Calls and History
You wanna know what happened after the call? You got this. First, use list_calls to get a summary log of multiple past calls—it’s your quick overview dashboard. If you need the deep dive, feed a specific Call ID into get_call_history. It pulls the detailed transcript and all the metadata for that one conversation.
Keep tabs on your account itself by calling get_user to grab basic details about your connected Thoughtly user account. This keeps everything synced up and tells you if your connection is solid.
The Bottom Line
This isn't just a list of tools; it’s a complete communication loop. Your agent handles the planning, the calling (trigger_call), the data gathering (using get_contact or list_contacts), and the record keeping (with create_contact and update_contact). It's the whole damn workflow, ready to go when you punch in a simple text command.
019dd174-20d9-73a8-b43a-e8ec60d994d8 
How Thoughtly MCP Works
- 1 Subscribe to this server and provide your Thoughtly API Token and Team ID.
- 2 Your text-based AI client issues a command, invoking a specific tool (e.g.,
trigger_call). - 3 Thoughtly executes the action—whether it's calling a number or updating a contact record—and returns structured data to your agent.
The bottom line is: your AI client becomes the dispatcher, ordering voice agents to interact with the real world using natural language prompts.
Who Is Thoughtly MCP For?
Sales Development Reps who need to qualify leads without manual data entry. Customer Support Engineers tired of switching between CRM and call log dashboards. Operations Managers running complex, multi-step communication campaigns that require both text follow-up and voice calls.
Automates outbound qualification by using trigger_call to engage new leads instantly.
Quickly pulls conversation context and transcripts for past interactions using get_call_history without opening the main dashboard.
Orchestrates multi-channel workflows by first running list_contacts, then triggering calls, and finally updating status with update_contact.
What Changes When You Connect
Real-World Use Cases
Cold Outreach Qualification
A new lead comes in. Instead of manually calling, the agent first uses list_contacts to ensure the number is registered. Then, it runs trigger_call, using the 'SDR Persona' to execute a scripted qualification call, solving the immediate need for contact engagement.
Post-Call Summary
A sales rep gets back from calls. They don't have time to write notes. The agent pulls the get_call_history for Call ID CAL-554, extracts key action items (e.g., 'Booked 2 PM Tuesday'), and uses that text to draft a follow-up email.
Cleanup Old Leads
The ops team finds old leads with bad data. They run list_contacts to see the list, then use get_contact on each one they want gone, and finally execute delete_contact, keeping the CRM clean.
Checking User Credentials
A manager needs to verify if a new user exists in Thoughtly before training them. The agent runs get_user with the employee ID, confirming access rights and basic profile details instantly.
The Tradeoffs
Creating duplicates
Trying to add a contact using create_contact when you already know the contact exists. This results in two records, one correct and one junk data.
→
Before running create_contact, always check if the record exists by calling get_contact. If it returns data, use that ID instead of creating a duplicate.
Calling without context
Running trigger_call with no specified agent persona or contact number. The call fails because the system doesn't know who to dial or what script to run.
→
First, list available agents using list_interviews. Then, use the returned ID in your prompt when calling trigger_call.
Ignoring history
Telling a user they need to call someone without checking what was said last week. They rehash old information and waste both time and effort.
→
Always run list_calls first, find the relevant Call ID, and then use get_call_history to pull the accurate transcript into your current context.
When It Fits, When It Doesn't
Use this server if your core business process involves turning text instructions into actual phone conversations or managing CRM data outside of a dedicated dashboard. You need an agent that can act in the real world using voice and structured records.
Don't use it if you just need to read static documents, analyze internal codebases, or perform calculations on spreadsheets; those are better handled by general LLM capabilities alone. If your goal is only to manage data structures (like JSON validation), look for a dedicated schema-checking tool instead of the full communication stack.
The critical boundary here is action: if you need the agent to do something—dial, create, update—then Thoughtly is required. For pure information retrieval or text processing, stick with your standard LLM setup.
Common Questions About Thoughtly MCP
How do I check which voice agents are available using Thoughtly MCP Server? +
Run the list_interviews tool. This returns a list of all active AI Voice Agents, giving you their unique IDs and names so you know exactly what persona to deploy.
Can I update contact information using Thoughtly MCP Server? +
Yes, use update_contact. You need the existing Contact ID and the specific field(s) you want to change (e.g., company size or status). It modifies only what you tell it to.
What is the difference between list_calls and get_call_history? +
list_calls gives you a high-level summary of many past calls (the log view). get_call_history requires a specific Call ID and returns the full, detailed transcript for that single conversation.
Do I need to run get_contact before trigger_call? +
While not mandatory, it's best practice. Running get_contact first confirms the contact details are current and validates that the ID you plan to use for trigger_call is accurate.
Can I mass delete contacts with Thoughtly MCP Server? +
The server provides a delete_contact tool, which handles single deletions. For bulk removal, you must first run list_contacts to get all IDs and then loop through them manually.
When I use list_contacts, what credentials must I provide to Thoughtly MCP Server? +
You need your active Thoughtly API Token and Team ID. These tokens authenticate your AI client and grant it the necessary permissions to read or write contact data via the server.
What metadata does get_call_history provide besides the conversation transcript? +
It returns comprehensive call metadata, including connection timestamps, total duration, and whether the call was marked as successful or failed. This lets you filter records based on outcome.
If I run trigger_call, how do I track the status of that outbound call? +
The server immediately queues the call and returns a unique Call ID. You must use this specific ID with get_call_history to monitor its real-time progress and retrieve the transcript later.
Can my text AI agent command a voice agent to call a specific person? +
Yes! Use the trigger_call tool by providing the Interview ID (Voice Agent) and the Contact ID. The voice AI will immediately execute the outbound call.
Is it possible to read the transcript of what the Voice AI discussed during a call? +
Absolutely. By calling the get_call_history tool with a specific Call ID, your agent can extract the entire conversational transcript and analyze the outcome without leaving the interface.
Do I need to manually register contacts in the Thoughtly dashboard before calling them? +
No. The integration provides the create_contact tool, allowing your AI to dynamically add new leads to the CRM and immediately initiate a phone call in one fluid workflow.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.