Connect SoScripted to Your AI Agent
Use the Model Context Protocol (MCP) to give your AI assistant the ability to transcribe videos, search your library, and manage your account — all through natural conversation.
What is MCP?
The Model Context Protocol is an open standard that lets AI assistants connect to external services. Instead of copying and pasting URLs or writing API calls, your AI agent can use SoScripted tools directly.
For example, you can tell Claude: "Transcribe this YouTube video and save it to my Research collection" — and it will handle everything automatically.
Supported AI Agents
Any application that supports the MCP standard can connect to SoScripted.
Setup Guide
Get your API key
Sign in to SoScripted and go to the API Keys page in your dashboard. Click Generate API Key and copy your key. It starts with sk_.
Add the configuration
Add the following to your MCP settings file. Replace sk_your_api_key with the key you copied.
OpenClaw / Claude Code / Cursor / Windsurf
{
"mcpServers": {
"soscripted": {
"type": "http",
"url": "https://soscripted.com/api/mcp",
"headers": {
"Authorization": "Bearer sk_your_api_key"
}
}
}
}Start using it
Restart your AI agent and you're ready to go. Try saying: "Transcribe this YouTube video: https://youtube.com/watch?v=..."
Example Prompts
Once connected, you can ask your AI agent things like:
Available Tools
Your AI agent gets access to 15 tools organized in 5 categories:
Transcription
transcribe
Transcribe a video URL from any of 7 supported platforms. Returns the transcript in your chosen format (txt, srt, vtt, json, timestamps, md). Costs 1 credit.
get_credits
Check your current credit balance and plan.
Library
search_library
Full-text search across all your saved transcripts. Filter by collection, paginate results.
get_transcript
Retrieve a saved transcript by ID in any output format.
save_transcript
Save a completed transcription job to your library, optionally into a collection.
Collections
list_collections
List all your collections with transcript counts.
create_collection
Create a new collection to organize transcripts.
delete_collection
Delete a collection. Transcripts inside are kept but unlinked.
Channel Monitors
list_monitors
List all your channel monitors with stats.
create_monitor
Auto-transcribe new videos from a YouTube channel. Filter by keywords, set monthly budget.
update_monitor
Update monitor settings (keywords, budget, active/inactive).
delete_monitor
Remove a channel monitor.
get_monitor_transcripts
Get recent transcripts from a monitored channel.
Batch Import
batch_transcribe
Transcribe all videos from a YouTube channel or playlist in one go.
get_batch_status
Check the progress of a batch import job.
Output Formats
Request transcripts in any of these formats:
| Format | Description |
|---|---|
| txt | Plain text without timestamps |
| srt | SubRip subtitle format |
| vtt | WebVTT subtitle format |
| json | Structured JSON with timestamps |
| timestamps | Text with inline timestamps |
| md | Markdown with metadata header |
Supported Platforms
Credits & Pricing
Each transcription costs 1 credit. If you transcribe the same video again, it hits the cache and still costs only 1 credit regardless of video length.
New accounts come with 3 free credits. Need more? Check our pricing plans.
Related Guides
Ready to connect?
Get your API key and start transcribing from your AI agent in under a minute.