Rev AI for AI agents
Rev AI converts speech to text, generating transcripts, captions, and subtitles for audio and video content using automated speech recognition. 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 Rev AI.
Check a transcription job
Get a job's status before pulling the finished transcript for a recording.
Pull captions for a video
Get captions for a completed job so they can be added to a published video.
Confirm before deleting a job
Get a job's details first, then delete it only after confirming the transcript is no longer needed.
36 Rev AI actions, graded by risk.
Every Rev AI action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
22Look things up. Allowed by default.
- rev_ai.get_accountGet Account
- rev_ai.get_captionsGet Captions
- rev_ai.get_job_by_idGet Job by ID
- rev_ai.get_topic_jobGet Topic Job
- rev_ai.list_topic_jobsList Topic Extraction Jobs
- rev_ai.get_list_of_jobsGet List of Jobs
- rev_ai.get_alignment_jobGet Alignment Job
- rev_ai.get_sentiment_jobGet Sentiment Job
- rev_ai.get_language_id_jobGet Language ID Job
- rev_ai.list_alignment_jobsList Alignment Jobs
- rev_ai.list_sentiment_jobsList Sentiment Analysis Jobs
- rev_ai.get_sentiment_resultGet Sentiment Analysis Result
- rev_ai.get_transcript_by_idGet Transcript by ID
- rev_ai.list_language_id_jobsList Language Identification Jobs
- rev_ai.get_language_id_resultGet Language Identification Result
- rev_ai.get_transcript_summaryGet Transcript Summary
write
8Create and change things. Allow, or ask the user first.
- rev_ai.create_rtmp_streamCreate RTMP Stream
- rev_ai.submit_alignment_jobSubmit Alignment Job
- rev_ai.submit_language_id_jobSubmit Language ID Job
- rev_ai.submit_custom_vocabularySubmit Custom Vocabulary
- rev_ai.submit_transcription_jobSubmit Transcription Job
- rev_ai.start_stream_transcriptionStart Stream Transcription
- rev_ai.submit_topic_extraction_jobSubmit Topic Extraction Job
- rev_ai.submit_sentiment_analysis_jobSubmit Sentiment Analysis Job
destructive
6Delete, cancel or archive. Ask first, or deny outright.
- rev_ai.delete_job_by_idDelete Job By ID
- rev_ai.delete_alignment_jobDelete Alignment Job
- rev_ai.delete_sentiment_jobDelete Sentiment Analysis Job
- rev_ai.delete_language_id_jobDelete Language ID Job
- rev_ai.delete_custom_vocabularyDelete Custom Vocabulary
- rev_ai.delete_topic_extraction_jobDelete Topic Extraction Job
Rev AI in three steps.
- 01Your users connect Rev AIThey add their Rev AI api key on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, writes like “create RTMP Stream” can wait for the user, and “delete Job By ID” can be denied outright.
- 03Any agent can actYour agent calls Rev AI through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('rev_ai', { read: 'allow', write: 'ask', // create_rtmp_stream destructive: 'deny', // delete_job_by_id }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How Rev AI connects.
Users add their Rev AI 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 Rev AI 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 Rev AI from any agent.
Rev AI and Arc0, answered.
Can I use Rev AI with Claude, ChatGPT or Cursor?
Yes. Connect Rev AI 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 Rev AI actions you allow.
How do users connect Rev AI?
Users add their Rev AI 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 Rev AI actions can my agent take?
36 in total: 22 read, 8 write and 6 destructive, such as “create RTMP Stream”. Your policies decide which of them each agent may call.
Can I stop my agent from deleting things in Rev AI?
Yes. Actions like “delete Job By ID” 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 Rev AI too?
Yes. The same Rev AI connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug Rev AI into your agent.
Your users connect Rev AI once, under your brand. Your agent gets 36 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan