AssemblyAI for AI agents
AssemblyAI provides speech-to-text transcription and audio intelligence through an API, used by developers to turn recordings and calls into searchable text and summaries. Connect it once through Arc0, and your agent, or Claude, ChatGPT and Cursor, can use it through one MCP endpoint, limited to what each user approved.
What agents do in AssemblyAI.
Transcribe an uploaded audio file
Submit an audio file for transcription and retrieve the resulting text once it is done.
Search within a transcript
Search a completed transcript for a specific word or phrase mentioned in it.
Confirm before deleting transcript content
Deleting a transcript's content is permanent and should be confirmed before it runs.
9 AssemblyAI actions, graded by risk.
Every AssemblyAI action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
6Look things up. Allowed by default.
- assemblyai.get_transcriptGet Transcript
- assemblyai.list_transcriptsList Transcripts
- assemblyai.search_transcriptSearch Transcript
- assemblyai.get_redacted_audioGet Redacted Audio
- assemblyai.get_transcript_segmentsGet Transcript Segments
- assemblyai.export_subtitlesExport Subtitles
write
2Create and change things. Allow, or ask the user first.
- assemblyai.upload_mediaUpload Media
- assemblyai.submit_transcriptSubmit Transcript
destructive
1Delete, cancel or archive. Ask first, or deny outright.
- assemblyai.delete_transcriptDelete Transcript Content
AssemblyAI in three steps.
- 01Your users connect AssemblyAIThey add their AssemblyAI api key on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, writes like “upload Media” can wait for the user, and “delete Transcript Content” can be denied outright.
- 03Any agent can actYour agent calls AssemblyAI through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('assemblyai', { read: 'allow', write: 'ask', // upload_media destructive: 'deny', // delete_transcript }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How AssemblyAI connects.
Users add their AssemblyAI api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
The same AssemblyAI connection serves your agent over MCP and your own backend over REST and the proxy, so a user connects once. How Arc0 handles credentials →
- AUTH
- API key
- CREDENTIALS
- Per-tenant encrypted vault
- MODEL SEES
- Results only, never credentials
- AUDIT LOG
- Every call, on every plan
Use AssemblyAI from any agent.
AssemblyAI and Arc0, answered.
Can I use AssemblyAI with Claude, ChatGPT or Cursor?
Yes. Connect AssemblyAI to Arc0 once, then add your Arc0 MCP URL to Claude, ChatGPT, Cursor, Claude Code or any other remote-MCP client. Each assistant only gets the AssemblyAI actions you allow.
How do users connect AssemblyAI?
Users add their AssemblyAI api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
Which AssemblyAI actions can my agent take?
9 in total: 6 read, 2 write and 1 destructive, such as “upload Media”. Your policies decide which of them each agent may call.
Can I stop my agent from deleting things in AssemblyAI?
Yes. Actions like “delete Transcript Content” are graded destructive. Set destructive actions to deny, or to ask so the user approves each one, and blocked calls still show up on the audit log.
Can my own backend call AssemblyAI too?
Yes. The same AssemblyAI connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug AssemblyAI into your agent.
Your users connect AssemblyAI once, under your brand. Your agent gets 9 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan