Get started

Human Approval - HITL Platform for AI agents

Human Approval (HITL.sh) adds human review checkpoints to AI agents and automations, letting a workflow pause until a person approves it. 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 Human Approval - HITL PlatformSee the 10 actions ↓
AI and modelsAPI keyMCP + RESThitl.sh
AUDIT LOG · HUMAN APPROVAL - HITL PLATFORMPOLICY: acme-support
09:41:07 · claude · u_8f2read
human_approval_hitl_platform.list_loops
List Loops✓ allowed · 212ms
09:41:08 · claude · u_8f2read
human_approval_hitl_platform.list_loop_members
List Loop Members✓ allowed · 164ms
09:41:09 · claude · u_8f2write
human_approval_hitl_platform.create_loop
Create Loop✓ approved · approved by user
09:41:12 · claude · u_8f2destructive
human_approval_hitl_platform.delete_loop
Delete Loop✕ blocked · policy: deny
EVERY HUMAN APPROVAL - HITL PLATFORM CALL, ON THE RECORD
01 · USE CASES

What agents do in Human Approval - HITL Platform.

01

Request approval for an action

Create an approval request before an automation takes a sensitive step, and wait for a human decision.

02

Check a pending request's status

Poll an approval request to see whether a reviewer has responded yet.

03

Cancel a stale request

Cancel an approval request that is no longer relevant, after confirming with the requester.

02 · ACTIONS

10 Human Approval - HITL Platform actions, graded by risk.

Every Human Approval - HITL Platform action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.

read

5

Look things up. Allowed by default.

  • human_approval_hitl_platform.list_loops
    List Loops
  • human_approval_hitl_platform.list_loop_members
    List Loop Members
  • human_approval_hitl_platform.get_account_status
    Get Account Status
  • human_approval_hitl_platform.get_approval_request
    Get Approval Request
  • human_approval_hitl_platform.list_approval_requests
    List Approval Requests

write

3

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

  • human_approval_hitl_platform.create_loop
    Create Loop
  • human_approval_hitl_platform.update_loop
    Update Loop
  • human_approval_hitl_platform.create_approval_request
    Create Approval Request

destructive

2

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

  • human_approval_hitl_platform.delete_loop
    Delete Loop
  • human_approval_hitl_platform.cancel_approval_request
    Cancel Approval Request
03 · HOW IT WORKS

Human Approval - HITL Platform in three steps.

  1. 01Your users connect Human Approval - HITL PlatformThey add their Human Approval - HITL Platform api key on Arc0 Connect, under your brand. It goes straight into the vault.
  2. 02You set the rulesReads run, writes like “create Loop” can wait for the user, and “delete Loop” can be denied outright.
  3. 03Any agent can actYour agent calls Human Approval - HITL Platform 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('human_approval_hitl_platform', {
  read: 'allow',
  write: 'ask',        // create_loop
  destructive: 'deny',  // delete_loop
})

# 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 Human Approval - HITL Platform connects.

Users add their Human Approval - HITL Platform 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 Human Approval - HITL Platform 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 Human Approval - HITL Platform 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

Human Approval - HITL Platform and Arc0, answered.

Q01

Can I use Human Approval - HITL Platform with Claude, ChatGPT or Cursor?

Yes. Connect Human Approval - HITL Platform 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 Human Approval - HITL Platform actions you allow.

Q02

How do users connect Human Approval - HITL Platform?

Users add their Human Approval - HITL Platform 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 Human Approval - HITL Platform actions can my agent take?

10 in total: 5 read, 3 write and 2 destructive, such as “create Loop”. Your policies decide which of them each agent may call.

Q04

Can I stop my agent from deleting things in Human Approval - HITL Platform?

Yes. Actions like “delete Loop” 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 Human Approval - HITL Platform too?

Yes. The same Human Approval - HITL Platform connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.

Get started

Plug Human Approval - HITL Platform into your agent.

Your users connect Human Approval - HITL Platform once, under your brand. Your agent gets 10 actions behind your policies, with every call on the record.

Connect Human Approval - HITL PlatformBrowse all apps →

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