ReelScribe MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| REELSCRIBE_API_KEY | Yes | Your ReelScribe API key, generated from your API settings at https://reelscribe.app/api (e.g. "rs_your_api_key_here"). Required for the local stdio server. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| transcribe_videoA | Submit a video URL for transcription and wait for the result. Supports Instagram Reels/Posts, TikTok videos, YouTube videos/Shorts, and Facebook videos/Reels. New jobs cost 1 credit. Existing matching transcripts may be reused without another charge. There is no storage quota. |
| get_transcriptionA | Get a transcription by its ID or requestId. Returns the full transcription text and any extracted comments when completed, or status/error info if still processing or failed. |
| list_transcriptionsA | List ONE PAGE of your transcriptions, newest first (default 20, max 200). The API does not report a library total. |
| search_transcriptionsA | Find transcriptions whose original video URL contains the given string. Returns the newest match in full, plus ids of any other matches. Useful for checking whether a video has already been transcribed. The search covers a bounded window of your most recent transcriptions and returns a capped number of matches, so |
| get_creditsA | Check your ReelScribe credit balance, subscription tier, how many completed transcriptions are stored, and how long they are kept. Each transcription costs 1 credit. There is no storage quota — a library is bounded by time (see |
| validate_urlA | Check if a URL is a supported video platform (Instagram, TikTok, YouTube, Facebook) without submitting for transcription. Returns the detected platform and normalized URL. |
| transcribe_bulkA | Submit up to 1,000 video URLs as a batch. The account tier sets the actual batch limit; each new video costs 1 credit. Existing transcripts are reused. Returns a bulk job ID. Requires an API key or additional batch transcription consent. |
| transcribe_instagram_channelA | Transcribe up to 50 reels from an Instagram channel, optionally after a date. Paid plans only. Reserves 1 credit per requested reel under the same refund rules as the website. Requires an API key or additional batch transcription consent. |
| list_jobsB | List one page of batch/channel jobs or read progress for a specific bulk job ID. Results belong to the connected account. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 9 tools
Each tool has a distinct action/resource: single vs bulk vs channel transcription are clearly scoped by their descriptions, and get/list/search are separable by purpose. Minor potential confusion between list_transcriptions and search_transcriptions, but descriptions make the boundary clear.
All tools follow a consistent snake_case verb_noun pattern (transcribe_video, get_transcription, list_transcriptions, search_transcriptions, validate_url, list_jobs, get_credits). transcribe_instagram_channel extends the pattern predictably.
Nine tools is well-scoped for a transcription service covering single, bulk, and channel workflows plus retrieval, validation, job listing, and billing. No tool appears redundant.
Core lifecycle is covered: submit (single/bulk/channel), retrieve (get/list/search), validate URLs, track jobs, and check credits. Minor gap: no delete/cleanup or export operation for stored transcriptions, though time-based retention may make deletion unnecessary.