Get started

Wit.ai for AI agents

Wit.ai builds natural language understanding for apps and devices, used by developers who need to parse intent and entities out of spoken or typed text. 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.

Connect Wit.aiSee the 31 actions ↓
AI and modelsAPI keyMCP + RESTwit.ai
AUDIT LOG · WIT.AIPOLICY: acme-support
09:41:07 · claude · u_8f2read
wit_ai.list_apps
List Wit.ai Apps✓ allowed · 212ms
09:41:08 · claude · u_8f2read
wit_ai.get_app
Get App Details✓ allowed · 164ms
09:41:09 · claude · u_8f2write
wit_ai.create_app
Create Wit.ai App✓ approved · approved by user
09:41:12 · claude · u_8f2destructive
wit_ai.delete_app
Delete App✕ blocked · policy: deny
EVERY WIT.AI CALL, ON THE RECORD
01 · USE CASES

What agents do in Wit.ai.

01

Get a message's parsed intent

Send text and get back the detected intent and entities, a read-only inference.

02

List an app's intents

Check the configured intents in an app before adding a new one.

03

Confirm before deleting an intent

Check with the user before deleting an intent, since anything built on it would stop matching.

02 · ACTIONS

31 Wit.ai actions, graded by risk.

Every Wit.ai action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.

read

15

Look things up. Allowed by default.

  • wit_ai.get_app
    Get App Details
  • wit_ai.get_trait
    Get Trait Details
  • wit_ai.get_voice
    Get Voice Details
  • wit_ai.list_apps
    List Wit.ai Apps
  • wit_ai.get_entity
    Get Entity Details
  • wit_ai.get_intent
    Get Intent Details
  • wit_ai.get_traits
    List Traits
  • wit_ai.get_intents
    Get Intents
  • wit_ai.get_message
    Wit.ai Get Message
  • wit_ai.list_voices
    List Voices
  • wit_ai.list_app_tags
    List App Tags
  • wit_ai.list_entities
    List Entities
  • wit_ai.list_utterances
    List Utterances
  • wit_ai.export_app
    Export App Data
  • wit_ai.detect_language
    Wit.ai Detect Language

write

9

Create and change things. Allow, or ask the user first.

  • wit_ai.create_app
    Create Wit.ai App
  • wit_ai.create_trait
    Create Wit.ai Trait
  • wit_ai.create_entity
    Create Wit.ai Entity
  • wit_ai.create_intent
    Create Wit.ai Intent
  • wit_ai.add_trait_value
    Add Value to Trait
  • wit_ai.create_utterances
    Create Wit.ai Training Utterances
  • wit_ai.add_entity_keyword
    Add Entity Keyword
  • wit_ai.add_keyword_synonym
    Add Keyword Synonym
  • wit_ai.put_app
    Update Wit.ai App

destructive

7

Delete, cancel or archive. Ask first, or deny outright.

  • wit_ai.delete_app
    Delete App
  • wit_ai.delete_entity
    Delete Entity
  • wit_ai.delete_intent
    Delete Intent
  • wit_ai.delete_utterances
    Delete Utterances
  • wit_ai.delete_entity_role
    Delete Entity Role
  • wit_ai.delete_entity_keyword
    Delete Entity Keyword
  • wit_ai.delete_keyword_synonym
    Delete Keyword Synonym
03 · HOW IT WORKS

Wit.ai in three steps.

  1. 01Your users connect Wit.aiThey add their Wit.ai api key on Arc0 Connect, under your brand. It goes straight into the vault.
  2. 02You set the rulesReads run, writes like “create Wit.ai App” can wait for the user, and “delete App” can be denied outright.
  3. 03Any agent can actYour agent calls Wit.ai through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
POLICY.TS
await arc0.policies.set('wit_ai', {
  read: 'allow',
  write: 'ask',        // create_app
  destructive: 'deny',  // delete_app
})

# Claude Code: the same connection, one URL
$ claude mcp add --transport http arc0 \
    https://mcp.arc0.ai/u/u_8f2
04 · AUTH AND DATA

How Wit.ai connects.

Users add their Wit.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 Wit.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
05 · WORKS WITH

Use Wit.ai from any agent.

Claude
ChatGPT
Cursor
Codex
VS Code
OpenAI Agents SDK
Claude Agent SDK
Vercel AI SDK
Mastra
LangGraph
06 · RELATED

More ai and models apps.

Perplexity AI9 actionsAffinda119 actionsAgent Mail5 actionsAI/ML API30 actionsAivoov1 actionsAll Images AI7 actionsAltText.ai10 actionsAltTextLab2 actions
ALL AI AND MODELS APPS →
07 · FAQ

Wit.ai and Arc0, answered.

Q01

Can I use Wit.ai with Claude, ChatGPT or Cursor?

Yes. Connect Wit.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 Wit.ai actions you allow.

Q02

How do users connect Wit.ai?

Users add their Wit.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.

Q03

Which Wit.ai actions can my agent take?

31 in total: 15 read, 9 write and 7 destructive, such as “create Wit.ai App”. Your policies decide which of them each agent may call.

Q04

Can I stop my agent from deleting things in Wit.ai?

Yes. Actions like “delete App” 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.

Q05

Can my own backend call Wit.ai too?

Yes. The same Wit.ai connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.

Get started

Plug Wit.ai into your agent.

Your users connect Wit.ai once, under your brand. Your agent gets 31 actions behind your policies, with every call on the record.

Connect Wit.aiBrowse all apps →

Free to build · MCP + REST · Audit log on every plan