Fliki MCP for AI. Turn Any Script Into a Video, Automatically.
Works with every AI agent you already use
…and any MCP-compatible client








How this MCP server connects to your AI agent
Fliki (AI Text-to-Video & Speech API) is your connector for creating full video content from simple text scripts. It lets you generate videos using customizable voices, set aspect ratios, and add background music, all without manual editing.
You can also check on the status of large batches and explore a list of available global AI voices to pick the perfect accent or language.
What AI agents can do with Fliki (AI Text-to-Video & Speech API) Automation
Generate video
Creates a new video file using the text script you provide and optional voice, aspect ratio, and music settings.
Get video
Retrieves the current status of a generation job and provides a direct download link if the video is finished.
List voices
Returns an organized list detailing all available AI voices and which languages they can speak.
Write a script and generate an entire video file, selecting voices, background music, and the required aspect ratio.
Check if a requested video is queued, processing, or ready for download, getting a public link when it's done.
Pull a list of every voice option and the languages they support so you can pick the right tone for your content.
Ask an AI about this
Waiting for input…
What AI agents can do with Fliki (AI Text-to-Video & Speech API) - 3 Tools
These three tools let you manage the entire video creation lifecycle: listing voices, starting generation, and retrieving the final file.
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 Fliki (AI Text-to-Video & Speech API) on VinkiusGenerate Video
Creates a new video file using the text script you provide and optional voice, aspect ratio, and music settings.
Get Video
Retrieves the current status of a generation job and provides a direct download link...
List Voices
Returns an organized list detailing all available AI voices and which languages they...
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.
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 Fliki (AI Text-to-Video & Speech API), 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
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
Built on the Model Context Protocol (MCP) for 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 3 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.
Video production used to be a multi-day nightmare.
Today, making a simple video clip means coordinating three people: the scriptwriter, the voice actor, and the editor. You write the copy, send it out for professional voice recording, then you spend hours in Premiere Pro or CapCut trimming silence, syncing music, and adjusting colors. The sheer friction of that manual process kills momentum.
With this MCP connector, your agent takes over those three roles instantly. You supply the script and tell us what tone and format you need. The system generates a finished video file with voices and background tracks, letting you get to the final product in minutes instead of days.
Generating Video Content via Fliki (AI Text-to-Video & Speech API)
You eliminate the need for external voice talent booking, studio time, and complex post-production software. You don't have to manage multiple files—no separate audio track, no background music file, just one complete video asset.
The difference now is pure speed. Instead of starting a tedious, multi-step project flow, you initiate the entire pipeline with your agent. It handles the generation and provides clear status updates until the link is ready.
What your AI can actually do with this
Turn raw text into finished video content using this MCP connector. Your agent handles the whole process: from picking the right voice to generating the final, shareable file. You give it a script; you tell it if you need a vertical clip for Instagram or a wide format for YouTube.
It figures out the voices and settings, then it builds the video, complete with background music tracks. If your project requires complex multimedia production, Vinkius's catalog makes connecting Fliki easy. Once the task is running, you don't have to wait; you can check back in later to see if it finished or if there are any issues.
019e5d1b-5d02-71ee-b7d1-e1b59ae3f620 Here's how it actually works
The bottom line is: you define the content details once, and your agent manages the multi-step process of creation and retrieval.
First, use the list_voices tool to see what voices and languages are available.
Next, call the generate_video tool, providing your script along with specific parameters like voice name, aspect ratio, and background music preference.
Finally, wait a short time, then run get_video using the provided task ID to check if the video is ready. If it is, you get the direct download link.
Who is this actually for?
Content strategists and marketing managers who hate spending hours coordinating video production. This MCP helps you run entire campaigns—from script to finished clip—without ever opening a dedicated video editor.
Needs to quickly generate 20 different promotional clips for various social platforms (TikTok, Instagram Story) using the same core message but adjusting aspect ratios and voices.
Writes chapter scripts and needs to automatically turn them into narrated video modules with consistent branding and voiceover talent.
Needs to rapidly prototype client pitches, generating multiple versions of videos for testing different tones or accents before recording any human talent.
What Changes When You Connect
Stop doing manual video editing. You simply hand off the script and let your agent handle voice selection and background music integration using generate_video.
Need multiple formats? Use one prompt to generate clips in different aspect ratios (16:9, 9:16, 1:1) for YouTube, TikTok, or Instagram stories instantly.
Worried about batch jobs failing? After calling generate_video, use get_video with the task ID. This lets you track status and download links in a systematic way.
Find the perfect voice every time. Before generating, run list_voices. You get access to all available accents and languages to match your target audience exactly.
Cut down on production time from days to minutes. Your agent acts as both the video editor and the studio director.
See it in action
Need a rapid series of social ads
A marketing team needs five variations of a product ad for different platforms (Instagram Reel, YouTube Short). They use list_voices to pick three tones, then call generate_video five times with the same script but varying the aspect ratio and voice parameters.
Publishing an e-book chapter
An author writes a chapter transcript. Instead of hiring a narrator and editor, they use generate_video to convert the text immediately into a polished video module for their online course.
Checking on a large campaign batch
After submitting 50 videos via generate_video, an agent uses get_video repeatedly. It confirms which videos are still processing and alerts the user only when all download links are active.
Localizing global content
A company needs to translate a message for markets in Brazil, Spain, and France. They use list_voices to verify local accents (Portuguese, Spanish, French) before generating the video using generate_video.
The honest tradeoffs
Assuming a voice is available
Trying to generate a video for 'Japanese accent' without checking first. The process fails halfway through because the specific accent isn't in the system.
Always start by running list_voices. This confirms if the language or accent you need is supported before spending time and resources generating the content.
Ignoring job status
Calling generate_video and then assuming the file is ready immediately. The system just accepted the request, but it takes minutes to render.
After calling generate_video, you must use get_video. This tool confirms the task ID's current status (queued, processing, or completed) before you try to download anything.
Forgetting aspect ratio
Generating a video for TikTok using default settings, only to find it’s in 16:9 format and looks wrong on the mobile feed.
When calling generate_video, always specify the required aspect ratio (e.g., 9:16) directly in the parameters.
When It Fits, When It Doesn't
Use this MCP if your primary need is transforming existing text into polished, narrated video content without complex filming or editing. This connector excels at structured output: script -> voiceover + music -> final MP4. Don't use it if you need to record live interviews, film original footage, or handle advanced graphical overlays that go beyond basic aspect ratios and backgrounds. If your goal is simply managing a database of video assets or coordinating multiple external services (like payment processing), look for a general workflow automation MCP instead. This tool is purely focused on content creation from text.
Questions you might have
How do I find out what voices are available using list_voices? +
Run list_voices to get a full directory of all AI voices and their supported languages. This lets you pick the right accent (like American English) before generating.
What is the difference between generate_video and get_video? +
generate_video starts the task, sending your script to the system for rendering. get_video checks on that specific job ID to see if it's finished or still processing.
Can generate_video handle different platforms like TikTok? +
Yes. When calling generate_video, you specify the aspect ratio (like 9:16) so the video is perfectly formatted for that platform right out of the gate.
Is there a way to change the voice after I call generate_video? +
No. You must select the desired voice and language parameters before running generate_video. Changes require starting a completely new generation task.
What do I need to do if my call to generate_video fails? +
If the generation fails, it will return an explicit error code. Check the full response body for details. Common issues include using unsupported aspect ratios or having a script that exceeds character limits.
Can I use get_video to check on tasks from yesterday? +
Yes, the status tracking is persistent for all created tasks. You simply need the unique task ID and can query the link anytime to see if it's complete or still processing.
What are the guidelines regarding background music when I call generate_video? +
You must provide a valid identifier for the desired background track. The tool accepts standard asset IDs; you can list available options and select one specific track to include in your video.
How do I handle rate limits when calling generate_video? +
The MCP handles basic rate limiting, but for high volume use, monitor the API response headers. If you hit a limit, implement an exponential backoff strategy before retrying your generation calls.
How can I see which AI voices are available for my video? +
Use the list_voices tool. It will return a full list of available AI voices along with their supported languages, helping you choose the right one for the generate_video action.
Can I track the progress of a video I just started generating? +
Yes! After starting a generation, use the get_video tool with the unique video ID. It will tell you if the video is 'queued', 'processing', or 'completed'.
How do I specify the video format, like for Instagram Reels or YouTube? +
When using generate_video, you can provide the aspectRatio parameter. Use '9:16' for Reels/TikTok, '16:9' for YouTube, or '1:1' for square posts.
We've already built the connector for Fliki. Just plug in your AI agents and start using Vinkius.
No hosting. No infrastructure. No complex setup.
All 3 tools are live and waiting.
You're up and running in seconds.
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.