Yes, you can run a video testimonial program from ChatGPT or Claude. Vocal Video connects to Claude, ChatGPT and other AI agents through its MCP server, so they can request, review, edit and share customer videos for you, in a chat or on a schedule while you're away. It's included on Essential plans and up, and every video waits as a draft until you publish it.
If you already spend half your day in an AI assistant, your testimonials are probably sitting in another tab, with responses waiting to be watched and an edit that never quite gets done. Connecting your assistant to Vocal Video moves that work into the chat you already have open.
In this guide:
- What the Vocal Video MCP server is
- What an AI assistant can do in Vocal Video
- Five prompts to copy
- Running it on a schedule
- What stays with you
- How to connect
- MCP or API?
- FAQ
What is the Vocal Video MCP server?
The Vocal Video MCP server lets an AI assistant work inside your Vocal Video account. MCP (Model Context Protocol) is the open standard AI assistants use to connect to other apps. Add one URL to your assistant, sign in to Vocal Video, and it can use the same tools as Vera, the AI assistant built into Vocal Video.
It works with:
- Claude, on the web, in the desktop app and in Claude Code
- ChatGPT, once Developer mode is switched on
- Cursor and other coding agents
- Any assistant or agent that supports remote MCP servers, including agents that run on a schedule
The MCP server is included on Essential, Pro, Scale and Enterprise plans; see pricing.
What can an AI assistant do in Vocal Video?
An AI assistant connected to Vocal Video can handle the everyday work of a video testimonial program: setting up collectors and sending video requests, reading and ranking responses, building and editing videos, and sharing them. Ask in plain words.
| Job | Ask your assistant | What happens |
|---|---|---|
| Request videos | "Create a collector called Customer Stories with three questions about the results our customers have seen, then write a video request for it." | It builds the collector and writes the request email, then shows you a preview before anything sends. |
| Review responses | "Summarize what people said in my Pricing collector, and star the three strongest answers about ROI." | It reads every transcript, sums up the themes and stars the best answers. |
| Make videos | "Make a draft video from Dana's response using our One-minute Testimonial template." | It builds a finished draft from that response, in your template's design. |
| Edit videos | "Tighten the pauses in the Acme testimonial, remove the filler words, and make it vertical for Instagram Reels." | It edits the draft: shorter gaps, no filler words, a 9:16 format. |
| Share videos | "Create a gallery of our five newest customer videos and give me the embed code." | It builds the gallery and hands back code to paste on your site. |
| Check your account | "How many AI credits do we have left?" | It reads your account's usage. |
Your assistant picks which answers to use and writes the titles and request emails. It never rewrites what someone said, so every word a customer says on screen is their own.
Five prompts to copy
The biggest time savings come from asking for several steps at once. Copy these and swap in your own names.
1. Start a customer-story campaign.
Create a collector called Q1 Customer Stories with questions about the problem they had, why they chose us, and the result they've seen. Write a video request for it, and show me the preview before you send it to the people in this spreadsheet.
2. Turn this quarter's best answers into a highlight reel.
Look at every response in the Customer Stories collector from this quarter, star the five strongest answers about results, and build a highlight reel from them. Send me the link to the draft.
3. Clean up a batch of drafts.
For every draft video from the Customer Stories collector this month, tighten the pauses, remove the filler words, and fix the spelling of "Acme" in the captions.
4. Refresh your website's testimonial gallery.
Swap anything older than 90 days in our homepage gallery for our newest published videos, and keep the first video where it is.
5. Hand drafts off for review.
Send the three newest drafts to Sam for review, and show me the email first.
Top tip: Name things exactly as they appear in Vocal Video. "The Customer Stories collector" finds the right collector every time; "my testimonials" makes your assistant guess.
Can an AI agent run it on a schedule?
Yes. Scheduled agents, like Claude's scheduled tasks, use the same Vocal Video connection while you're away. Write the instructions once, and the agent runs them every morning or every week. Any always-on agent that can connect to a custom MCP server works the same way, including self-hosted ones like OpenClaw.
Three recipes to start with:
A Monday summary with drafts ready
Every Monday at 8am, find last week's new responses in all my collectors, star the strongest answer in each, make a draft video from each starred response, and send me a summary with links to the drafts.
A testimonial gallery that stays fresh
On the first of every month, swap anything older than 90 days in our homepage gallery for our newest published videos.
Follow-ups for quiet collectors
Every Friday, tell me which collectors got no new responses this week, and write a follow-up video request for each one. Include the previews in your summary, and don't send anything.
Tell your agent what it may send. In a chat, your assistant shows you a preview before any email goes out and waits for your OK. A scheduled task has no one to ask, so put the rule in its instructions: either "send the requests" or "show me the previews and don't send." Drafts are different: an agent can build and edit as many as you like, but only a person can publish.
When to use Zapier, webhooks or the API instead. A scheduled agent runs at the times you set. When something should happen the moment it occurs, like a deal closing in your CRM or a new response arriving, use Zapier (on any paid plan) or webhooks and the API (on Pro plans and up). They work well together: Zapier sends the video request when a deal closes, and your agent reviews the week's responses every Monday. Our guide to automating video testimonials covers that side.
What does the agent do on its own, and what stays with you?
Your AI assistant does the clicking in Vocal Video; you keep the decisions that matter. Here's how the connection is built:
- It works as you, in one workspace. Your assistant gets your role's permissions, in the workspace you chose when you connected. A reviewer's assistant can look through responses and star them, but can't change anything else.
- Emails wait for your OK. In a chat, you see every video request and review email before it goes out. On a schedule, the agent follows your instructions.
- It builds drafts; you publish. No video goes public until a person clicks Publish in Vocal Video.
- It never sees your password. You sign in on Vocal Video, and your assistant gets a short-lived token that renews itself.
- You can cut it off anytime. Disconnect it in Vocal Video's settings or from your assistant, and its access stops right away.
It all runs on the same ISO 27001 and ISO 27701 certified platform that more than 30,000 companies use for their customer and employee videos.
How do I connect my AI assistant to Vocal Video?
Connecting takes about two minutes and three steps:
- In Vocal Video, go to Settings > Connect AI Assistant (MCP) and copy your server URL.
- Add the URL to your assistant as a custom connector. In Claude, that's under Customize > Connectors; in ChatGPT, switch on Developer mode, then add it under Plugins.
- When the Vocal Video sign-in opens, log in, choose a workspace and click Allow.
For click-by-click setup in Claude, ChatGPT, Claude Code and Cursor, see MCP: Use Vocal Video From Claude, ChatGPT and Other AI Assistants in our help center. Developers can find the full tool reference in the developer docs.
MCP or API: which do I need?
Use the MCP server if a person or an AI agent is doing the work; use the API if your own software is. Both reach the same tools with the same permissions.
| Use | When | Plans |
|---|---|---|
| The MCP server | You're asking Claude, ChatGPT or another assistant, or running an AI agent on a schedule. Paste a URL and sign in. | Essential, Pro, Scale and Enterprise |
| The REST API | You're building your own integration, script or backend, with an API key. | Pro, Scale and Enterprise |
Frequently asked questions
Is there a video testimonial tool I can run from ChatGPT or Claude?
Yes. Vocal Video connects to ChatGPT, Claude and other AI assistants through its MCP server. Once connected, you can ask your assistant to send video requests, review responses, build and edit videos, and share them. Every video waits as a draft for you to publish. It's included on Essential plans and up.
Which Vocal Video plans include the MCP server?
The MCP server is included on Essential, Pro, Scale and Enterprise plans; Essential starts at $99/mo, billed annually. The REST API, for building your own integrations, is included on Pro, Scale and Enterprise.
Does using my AI assistant with Vocal Video use AI credits?
Not on its own. The conversation runs on your assistant's plan, and most of what your assistant does in Vocal Video is free. Two things spend Vocal Video AI credits, just as they do in the app: drafts built with an AI-directed template, and publishing videos with AI effects such as filler-word removal.
Can my AI assistant publish videos?
No. Your assistant can build and edit drafts in Vocal Video, but publishing is always a person's click.
How is this different from Vera?
Vera is the AI assistant built into Vocal Video. Connecting your own assistant gives you the same kind of help from Claude, ChatGPT or another assistant you already use, and lets an agent run tasks on a schedule.
Can my teammates use my connection?
No. Each person connects with their own Vocal Video login, so every assistant works with that person's permissions.
Let your assistant handle the busywork
A video testimonial program has always had two halves: the human part, where a customer tells their story, and the busywork around it. Connect your AI assistant to Vocal Video and the busywork moves into the chat you already use, or runs on a schedule while you're away. Every video still waits for your OK before it goes live.
Ready to try it? Sign up for a free Vocal Video account. Your free trial includes the MCP server. No credit card required.
