Skip to content

TranscriptLab MCP server

dev.transcriptlab/transcriptlabtranscriptlab.devstreamable-httpv1.0.0Scanned

Turn YouTube videos and audio or video files into searchable, exportable transcripts.

19 tools: 1 do something that can't be undone, 10 change things and 8 only read.

Tools: 0 delete, 1 irreversible, 10 write, 8 read.
  • Reads private data · 2
  • Reads text others write · 14
  • Can send data out · 7
  • 4 findings

Tools listed without sign-in. Remote posture

With Anlyon

1 tool here does something that can't be undone. Govern it.

  • Anlyon lists the tools and adopts none.
  • Each tool you adopt gets approvals and receipts.
  • Your agent never holds the key.
Govern this server →

A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.

What the TranscriptLab MCP server can do

The TranscriptLab MCP server is published by transcriptlab.dev, a domain the MCP Registry verified. It has 19 tools. It can do things that cannot be undone, such as sending, paying or publishing (transcriptlab_start_audio_transcription) and create or change data (10 tools: transcriptlab_get_transcript, transcriptlab_retrieve_transcript, transcriptlab_mentions, transcriptlab_open_audio and 6 more), but cannot delete data. 8 tools only read.

2 tools read private data (transcriptlab_read_audio_transcript and transcriptlab_get_speech_transcript); 14 tools bring in text other people can write (transcriptlab_get_transcript, transcriptlab_retrieve_transcript, transcriptlab_discover_captions and 11 more); 7 tools can send data out of your system (transcriptlab_get_transcript, transcriptlab_retrieve_transcript, transcriptlab_discover_captions and 4 more). On its own it has all three parts of a prompt-injection exfiltration path: private data, text others write, and a way out.

The fixed checks matched 4 findings (annotation-conflict). Each one comes with the text it matched.

Its remote endpoint lists its tools without sign-in.

Last scanned Nov 17, 2026, version 1.0.0.

Tools

Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.

19 tools

Can't be undone 1Claude Code: ask
  • transcriptlab_start_audio_transcriptionFinalize a signed upload and start speech-to-text on it.No findings

    Why it is in this group

    • The name "transcriptlab_start_audio_transcription" says "start": a word for a change.
    • The description's first verb is "finalize": a word for a change that another call does not undo.
    Description, as the server wrote it
    Finalize a signed upload and start speech-to-text on it. App-facing only: the uploader view calls it once the file is uploaded. Spends credits per audio minute.

    Annotations the server declares

    • title: Transcribe uploaded audio
    • readOnlyHint: false
    • destructiveHint: false
    • openWorldHint: false
Changes things 10Claude Code: ask
  • transcriptlab_begin_audio_uploadCreate a signed direct-upload session for an opened audio file.No findings

    Why it is in this group

    • The name "transcriptlab_begin_audio_upload" says "upload": a word for a change.
    Description, as the server wrote it
    Create a signed direct-upload session for an opened audio file. App-facing only: the uploader view calls it, not the model.

    Annotations the server declares

    • title: Begin audio upload
    • readOnlyHint: false
    • destructiveHint: false
    • openWorldHint: false
  • transcriptlab_get_transcriptGet a YouTube transcript from existing captions.Reads text others writeCan send data out1 finding

    Why it is in this group

    • The name "transcriptlab_get_transcript" says "get": a word for reading.
    • The server marks it as not read-only (readOnlyHint: false).

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • untrusted-inputThe input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.

    Findings

    • Warn

      The server marks "transcriptlab_get_transcript" as not read-only (readOnlyHint: false), and its words say it reads.

      Matched text

      annotations: {"readOnlyHint":false,"destructiveHint":false,"openWorldHint":true}; The name "transcriptlab_get_transcript" says "get": a word for reading.

      What this check looks for
    Description, as the server wrote it
    Get a YouTube transcript from existing captions. Use when the user gives a YouTube link or video ID and wants the transcript. Returns the transcript, or a job id when the request outlives the synchronous window: if a job comes back, poll transcriptlab_get_job_status, then read the text with transcriptlab_retrieve_transcript. The first delivery of each transcript variant can spend credits; later deliveries of that variant are free. Captions only: when the user asks which caption tracks exist use transcriptlab_discover_captions; when the video has no captions, or the source is an attached file or a direct media link, use transcriptlab_transcribe_file or transcriptlab_transcribe_media_url.

    Annotations the server declares

    • title: Get YouTube transcript
    • readOnlyHint: false
    • destructiveHint: false
    • openWorldHint: true
  • transcriptlab_mentionsComposer at-mention search over the account’s transcript library, used to build the @TranscriptLab picker.Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Composer at-mention search over the account’s transcript library, used to build the @TranscriptLab picker. App-facing only: the host calls it, it returns resource links rather than transcript text, and the model should not call it.

    Annotations the server declares

    • title: Transcript library
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: true
    • idempotentHint: true
  • transcriptlab_openOpen the TranscriptLab launcher view, where the user can work with transcripts visually and send one into the chat with "Discuss in chat".1 finding

    Why it is in this group

    • The name "transcriptlab_open" says "open": a word for a change.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    Findings

    • Warn

      The server marks "transcriptlab_open" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

      Matched text

      annotations: {"readOnlyHint":true,"destructiveHint":false,"openWorldHint":false}; The name "transcriptlab_open" says "open": a word for a change.

      What this check looks for
    Description, as the server wrote it
    Open the TranscriptLab launcher view, where the user can work with transcripts visually and send one into the chat with "Discuss in chat". Use tab "upload" when a file needs the uploader: the user wants to pick a local audio or video file, the attached file is too large or could not be read by transcriptlab_transcribe_file, or that upload failed. Use "library" to reopen earlier transcripts, and "link" (the default) to paste a YouTube link. Also offer it when the user wants a transcript but has given no link or file.

    Annotations the server declares

    • title: Open TranscriptLab
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • transcriptlab_open_audioEntry point the host invokes when the user opens an audio or video file with this plugin: opens the uploader view and starts speech-to-text.Reads text others writeNo findings

    Why it is in this group

    • The name "transcriptlab_open_audio" says "open": a word for a change.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Entry point the host invokes when the user opens an audio or video file with this plugin: opens the uploader view and starts speech-to-text. You do not call this directly; for a file attached to the conversation use transcriptlab_transcribe_file.

    Annotations the server declares

    • title: Transcribe an audio file
    • readOnlyHint: false
    • destructiveHint: false
    • openWorldHint: true
  • transcriptlab_retrieve_transcriptRead a YouTube transcript that is already published for this account, without starting a new fetch.Reads text others writeCan send data out1 finding

    Why it is in this group

    • The name "transcriptlab_retrieve_transcript" says "retrieve": a word for reading.
    • The server marks it as not read-only (readOnlyHint: false).

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • untrusted-inputThe input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.

    Findings

    • Warn

      The server marks "transcriptlab_retrieve_transcript" as not read-only (readOnlyHint: false), and its words say it reads.

      Matched text

      annotations: {"readOnlyHint":false,"destructiveHint":false,"openWorldHint":true}; The name "transcriptlab_retrieve_transcript" says "retrieve": a word for reading.

      What this check looks for
    Description, as the server wrote it
    Read a YouTube transcript that is already published for this account, without starting a new fetch. Use it after transcriptlab_get_job_status reports a job done, or to re-open a variant that was delivered before. For a first request use transcriptlab_get_transcript instead. Reading an already delivered variant is free; a never-before-delivered variant can spend credits like the original request.

    Annotations the server declares

    • title: Retrieve completed transcript
    • readOnlyHint: false
    • destructiveHint: false
    • openWorldHint: true
  • transcriptlab_submit_batchQueue a list of YouTube URLs as one batch job and return its job id.Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "transcriptlab_submit_batch" says "submit": a word for a change.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • outboundThe input takes a URL (input.urls): a request to a URL can carry data to whoever runs it.
    Description, as the server wrote it
    Queue a list of YouTube URLs as one batch job and return its job id. App-facing only. The plan caps how many URLs one job accepts.

    Annotations the server declares

    • title: Submit a batch transcript job
    • readOnlyHint: false
    • destructiveHint: false
    • openWorldHint: true
  • transcriptlab_transcribe_fileTranscribe an audio or video file the user attached to this conversation, using speech-to-text (mp3, m4a, wav, mp4, webm and similar).Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Transcribe an audio or video file the user attached to this conversation, using speech-to-text (mp3, m4a, wav, mp4, webm and similar). Use it for an attached media file; YouTube links use transcriptlab_get_transcript instead. Spends credits per audio minute. Attachments up to 300 MB work here; larger files (up to 2 GB) use the launcher upload view. Stores the file, starts transcription, and returns an upload_id: poll transcriptlab_get_speech_transcript with it, waiting a few seconds between tries and stopping after about ten. Do not call this again for the same file. If it fails or the file is too large, call transcriptlab_open with tab "upload".

    Annotations the server declares

    • title: Transcribe an attached audio or video file
    • readOnlyHint: false
    • destructiveHint: false
    • openWorldHint: true
  • transcriptlab_transcribe_media_urlTranscribe a public https link to an audio or video file with speech-to-text.Reads text others writeCan send data outNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
    Description, as the server wrote it
    Transcribe a public https link to an audio or video file with speech-to-text. Use it for a direct media URL (ending in .mp3, .m4a, .wav, .mp4, .webm and similar); not for YouTube or other web pages, which use transcriptlab_get_transcript. Spends credits per audio minute. Returns a media_id: poll transcriptlab_get_speech_transcript with it, waiting a few seconds between tries and stopping after about ten.

    Annotations the server declares

    • title: Transcribe a direct media link
    • readOnlyHint: false
    • destructiveHint: false
    • openWorldHint: true
Only reads 8Show 8 read-only tools
  • transcriptlab_discover_captionsList the caption languages and track types (manual vs generated) available for a YouTube video.Reads text others writeCan send data outNo findings

    Why it is in this group

    • The description's first verb is "list": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • untrusted-inputThe input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.

    Claude Code: ask, because it is tagged outbound.

    Description, as the server wrote it
    List the caption languages and track types (manual vs generated) available for a YouTube video. Use when the user asks which languages or subtitles a video has, or before transcriptlab_get_transcript when they want a specific language. Free in transcript credits, but subject to the account monthly discovery limit.

    Annotations the server declares

    • title: Discover caption tracks
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: true
  • transcriptlab_get_job_statusCheck whether a transcript job from transcriptlab_get_transcript has finished.Reads text others writeNo findings

    Why it is in this group

    • The name "transcriptlab_get_job_status" says "get": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Check whether a transcript job from transcriptlab_get_transcript has finished. Use it when that tool returned a job_id instead of a transcript. Returns status only, delivers no transcript, and spends no credits; once it reports done, read the text with transcriptlab_retrieve_transcript.

    Annotations the server declares

    • title: Check transcript job
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: true
    • idempotentHint: true
  • transcriptlab_get_speech_transcriptRead the speech-to-text transcript started by transcriptlab_transcribe_file (pass upload_id) or transcriptlab_transcribe_media_url (pass media_id).Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "transcriptlab_get_speech_transcript" says "get": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "read" and "file": it reads files.
    • untrusted-inputThe description's first sentence says "read" and "url": it reads web pages, which anyone can write.
    Description, as the server wrote it
    Read the speech-to-text transcript started by transcriptlab_transcribe_file (pass upload_id) or transcriptlab_transcribe_media_url (pass media_id). Pass exactly one id. While transcription runs it returns a pending result: wait a few seconds, back off, and stop after about ten tries; if it is still pending, tell the user the job is queued instead of looping. Reading does not start new work or add charges.

    Annotations the server declares

    • title: Read a speech-to-text transcript
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • transcriptlab_get_usageRead the account plan, remaining credits, and current-period usage.Reads text others writeNo findings

    Why it is in this group

    • The name "transcriptlab_get_usage" says "get": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Read the account plan, remaining credits, and current-period usage. Use when the user asks about cost, balance, or limits, or before a credit-spending request.

    Annotations the server declares

    • title: Check TranscriptLab usage
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: true
    • idempotentHint: true
  • transcriptlab_list_job_itemsReturn per-video results for a batch or channel job.Reads text others writeNo findings

    Why it is in this group

    • The name "transcriptlab_list_job_items" says "list": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Return per-video results for a batch or channel job. App-facing only. Page with the returned cursor.

    Annotations the server declares

    • title: List job items
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: true
  • transcriptlab_list_transcriptsList YouTube videos whose transcripts were already delivered to this account, most recent first.Reads text others writeNo findings

    Why it is in this group

    • The name "transcriptlab_list_transcripts" says "list": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    List YouTube videos whose transcripts were already delivered to this account, most recent first. Use it to find or reopen an earlier transcript. It fetches nothing and delivers no transcript. When more results exist, page with the returned cursor.

    Annotations the server declares

    • title: List transcript library
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: true
    • idempotentHint: true
  • transcriptlab_read_audio_transcriptRead the transcript of a file uploaded through the uploader view.Reads private dataNo findings

    Why it is in this group

    • The name "transcriptlab_read_audio_transcript" says "read": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "read" and "file": it reads files.
    Description, as the server wrote it
    Read the transcript of a file uploaded through the uploader view. A pending transcription reports 202; retry shortly. App-facing only.

    Annotations the server declares

    • title: Read uploaded-audio transcript
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • transcriptlab_search_transcriptSearch, quote, or slice a transcript already delivered to this account.Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "transcriptlab_search_transcript" says "search": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • untrusted-inputThe input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.

    Claude Code: ask, because it is tagged outbound.

    Description, as the server wrote it
    Search, quote, or slice a transcript already delivered to this account. Use for "find the part where...", keyword or regex lookup, or paginating a long transcript into chunks. It never starts a fetch and never charges transcript credits, so it only works on a video already cached here: deliver it first with transcriptlab_get_transcript if it is not. mode chooses the behavior: keyword matches terms, regex matches a pattern, chunk returns sequential sections.

    Annotations the server declares

    • title: Search a cached transcript
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: true
    • idempotentHint: true

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Exfil surface

Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.

196 paths in this server alone from private data, through text someone else wrote, to a way out. Denying 2 private-data tools breaks every one.

Reads private data2
Reads files, mail, messages, databases, secrets, calendars or repositories.
deny to break every path
{
  "permissions": {
    "deny": [
      "mcp__transcriptlab__transcriptlab_get_speech_transcript",
      "mcp__transcriptlab__transcriptlab_read_audio_transcript"
    ]
  }
}

Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.

Remote posture

What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.

Tools listed without sign-in

Looked at Nov 17, 2026, 15:19 UTC, unchanged since Nov 17, 2026. No credentials were sent and no tool was called. What is looked at.

Reachable
yes
HTTP status of an anonymous initialize
200
tools/list served without credentials
yes
Tools that change things, listed without credentials
11
Protocol version
2025-06-18
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
not observed
RFC 9728 protected resource metadata
https://api.transcriptlab.dev/.well-known/oauth-protected-resource/mcp
Authorization servers
https://api.transcriptlab.dev, https://clerk.transcriptlab.dev
Legacy HTTP+SSE transport
no
Dynamic client registration
yes
Client ID metadata documents
no
  • HighThe whole server

    The endpoint served tools/list without credentials, and the list has 11 tools that change things (1 irreversible, 10 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://api.transcriptlab.dev/mcp: tools/list returned 19 tools to an anonymous client, including transcriptlab_start_audio_transcription (irreversible), transcriptlab_get_transcript (write), transcriptlab_retrieve_transcript (write), transcriptlab_mentions (write), transcriptlab_open_audio (write) and 6 more

    What this check looks for

Findings

Matches of fixed checks, with the text they matched.

4 warn
  • The server marks "transcriptlab_get_transcript" as not read-only (readOnlyHint: false), and its words say it reads.

    Matched text

    annotations: {"readOnlyHint":false,"destructiveHint":false,"openWorldHint":true}; The name "transcriptlab_get_transcript" says "get": a word for reading.

    What this check looks for
  • The server marks "transcriptlab_retrieve_transcript" as not read-only (readOnlyHint: false), and its words say it reads.

    Matched text

    annotations: {"readOnlyHint":false,"destructiveHint":false,"openWorldHint":true}; The name "transcriptlab_retrieve_transcript" says "retrieve": a word for reading.

    What this check looks for
  • The server marks "transcriptlab_open" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

    Matched text

    annotations: {"readOnlyHint":true,"destructiveHint":false,"openWorldHint":false}; The name "transcriptlab_open" says "open": a word for a change.

    What this check looks for
  • The server marks "transcriptlab_export_link" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

    Matched text

    annotations: {"readOnlyHint":true,"destructiveHint":false,"openWorldHint":false}; The name "transcriptlab_export_link" says "link": a word for a change.

    What this check looks for

History

Changes to its tool definitions and to its remote endpoint, newest first.

No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.

Install

Pinned to version 1.0.0. Values in angle brackets are yours to fill in.

claude mcp add --transport http transcriptlab https://api.transcriptlab.dev/mcp
claude mcp add --transport http transcriptlab https://api.transcriptlab.dev/mcp

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__transcriptlab__transcriptlab_get_job_status",
      "mcp__transcriptlab__transcriptlab_get_usage",
      "mcp__transcriptlab__transcriptlab_list_transcripts",
      "mcp__transcriptlab__transcriptlab_read_audio_transcript",
      "mcp__transcriptlab__transcriptlab_get_speech_transcript",
      "mcp__transcriptlab__transcriptlab_list_job_items"
    ],
    "ask": [
      "mcp__transcriptlab__transcriptlab_get_transcript",
      "mcp__transcriptlab__transcriptlab_retrieve_transcript",
      "mcp__transcriptlab__transcriptlab_discover_captions",
      "mcp__transcriptlab__transcriptlab_search_transcript",
      "mcp__transcriptlab__transcriptlab_mentions",
      "mcp__transcriptlab__transcriptlab_open_audio",
      "mcp__transcriptlab__transcriptlab_begin_audio_upload",
      "mcp__transcriptlab__transcriptlab_start_audio_transcription",
      "mcp__transcriptlab__transcriptlab_open",
      "mcp__transcriptlab__transcriptlab_transcribe_file",
      "mcp__transcriptlab__transcriptlab_transcribe_media_url",
      "mcp__transcriptlab__transcriptlab_export_link",
      "mcp__transcriptlab__transcriptlab_submit_batch"
    ],
    "deny": []
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as transcriptlab. Review it before you use it.

Capability tags behind 16 rules
  • mcp__transcriptlab__transcriptlab_get_transcript untrusted-input, outbound
  • mcp__transcriptlab__transcriptlab_retrieve_transcript untrusted-input, outbound
  • mcp__transcriptlab__transcriptlab_discover_captions untrusted-input, outbound
  • mcp__transcriptlab__transcriptlab_search_transcript untrusted-input, outbound
  • mcp__transcriptlab__transcriptlab_get_job_status untrusted-input
  • mcp__transcriptlab__transcriptlab_get_usage untrusted-input
  • mcp__transcriptlab__transcriptlab_list_transcripts untrusted-input
  • mcp__transcriptlab__transcriptlab_mentions untrusted-input
  • mcp__transcriptlab__transcriptlab_open_audio untrusted-input
  • mcp__transcriptlab__transcriptlab_read_audio_transcript private-data
  • mcp__transcriptlab__transcriptlab_transcribe_file untrusted-input
  • mcp__transcriptlab__transcriptlab_transcribe_media_url untrusted-input, outbound
  • mcp__transcriptlab__transcriptlab_get_speech_transcript private-data, untrusted-input
  • mcp__transcriptlab__transcriptlab_export_link outbound
  • mcp__transcriptlab__transcriptlab_submit_batch untrusted-input, outbound
  • mcp__transcriptlab__transcriptlab_list_job_items untrusted-input

Pin

Keep the decision: fail CI when the tools change, follow every change, or show a badge.

Pin in CI

Fail a build when the tools you reviewed are not the tools the server serves today.

curl -s "https://radar.anlyon.com/api/v1/servers/dev.transcriptlab/transcriptlab/verify?hash=ad90a90e4e60b54fade14f0678afc95c65d90b9c286cabefba5990d2cac9aa26" | jq -e '.changed == false'

Pinned hash ad90a90e4e60, from Nov 17, 2026, 11:18 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/dev.transcriptlab/transcriptlab/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for dev.transcriptlab/transcriptlab
[![Anlyon Radar](https://radar.anlyon.com/badge/dev.transcriptlab/transcriptlab.svg)](https://radar.anlyon.com/servers/dev.transcriptlab/transcriptlab)

Source: MCP Registry · repository. Synced Nov 17, 2026, 11:17 UTC.

TranscriptLab MCP server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the TranscriptLab MCP server safe?

No scan can promise that. What its definitions show: 19 tools, 11 of which can change, send or delete something, and 4 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.

Can the TranscriptLab MCP server delete data?

No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.

Can the TranscriptLab MCP server do things that cannot be undone?

Yes. 1 tool can send, pay, publish or merge: transcriptlab_start_audio_transcription.

What tools does the TranscriptLab MCP server have?

19 tools: 1 irreversible, 10 write, 8 read. transcriptlab_start_audio_transcription, transcriptlab_get_transcript, transcriptlab_retrieve_transcript, transcriptlab_mentions, transcriptlab_open_audio, transcriptlab_begin_audio_upload, transcriptlab_open, transcriptlab_transcribe_file and 11 more.

Does the TranscriptLab MCP server need sign-in or OAuth?

Its remote endpoint listed its tools to an anonymous client, without sign-in.

Does the TranscriptLab MCP server send data to outside services?

7 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): transcriptlab_get_transcript, transcriptlab_retrieve_transcript, transcriptlab_discover_captions, transcriptlab_search_transcript, transcriptlab_transcribe_media_url, transcriptlab_export_link and 1 more.

Has the TranscriptLab MCP server changed its tools?

Not since Anlyon Radar started tracking it. Last scanned Nov 17, 2026.