Human Approval for AI Agents

Agents run wild.
mild.run
Keep it mild.

Before your AI acts on anything sensitive — refunds, emails, data deletion — mild.run intercepts and asks you first. No terminal, no SDK, no webhooks. Approve or block from your preferred channel in seconds.

Without mild.run — agents run wild
gate
With mild.run — every action approved
mild.run
approval gate
24 hrapproval window
1 lineto add to any agent
100%auditable decisions
< 5 minto set up
How it works

Three steps to run
agents mild.

01
Agent triggers a sensitive action

Your AI reaches a critical operation — a Stripe refund, an email blast, a record deletion. Instead of firing immediately, it pauses and calls mild.run.

02
mild.run intercepts and notifies

mild.run sends you a notification with full context: what the agent wants to do, why, and how much. The agent waits — up to 24 hours.

03
You decide. Always.

Tap Approve or Block from your phone. The agent acts only with your explicit sign-off. Every decision is logged and auditable.

What mild.run protects

Every sensitive action,
intercepted before it fires.

Stripe
Stripe
  • Refunds
  • Subscription cancellations
  • Customer deletions
  • Payouts
4 actions · Live
Gmail
Gmail
  • Send email
1 action · Live
Slack
Slack
  • Post message
1 action · Live
More integrations

HubSpot, Notion, custom actions, and much more — expanding fast.

Coming soon
Notification channels

Get notified
your way.

No new app to install. Approval requests are delivered to your preferred channel — approve or block with a single tap.

✉️EmailLive
TelegramTelegramLive
WhatsAppWhatsAppComing soon

More channels coming soon.

Get started →
M
mild.run
● Online — Approval Bot
⏸ Approval required
ActionStripe refund
Amount$240.00
Chargech_3Qx8a…
ReasonCustomer request
Expires in 23h 59m · mild.run
Pricing

Simple, transparent pricing.

Free
$0/month

Perfect for trying mild.run on personal projects.

  • 30 approvals per month
  • Up to 2 integrations
  • Email & Telegram notifications
  • 24 hr approval window
  • Full audit log
Get started free →
Most popular
Solo
$12/month

For developers running agents in production.

  • Unlimited approvals
  • All integrations
  • Email & Telegram notifications
  • 24 hr approval window
  • Full audit log
  • Priority support
Start for free →
FAQ

Frequently asked
questions.

What is human-in-the-loop for AI agents?

Human-in-the-loop means an AI agent pauses before a sensitive action — a refund, an email, a deletion — and waits for a person to explicitly approve or block it. mild.run adds this gate to any agent through one MCP tool call.

Which actions can mild.run gate?

mild.run currently gates Stripe refunds, subscription cancellations, customer deletions and payouts, Gmail sends, and Slack messages. More integrations are added regularly.

How does the agent get my decision?

mild.run notifies you by Telegram, email or Slack with full context on the requested action. You tap Approve or Block, and the agent receives that decision — waiting up to 24 hours before treating a non-response as blocked.

Does mild.run require changes to my agent's code?

No SDK or webhook server is required. mild.run is exposed as an MCP tool, so any MCP-compatible agent can call it directly to request approval.

Is mild.run free to use?

Yes. The Free plan includes 30 approvals per month across up to 2 integrations. The Solo plan is $12/month for unlimited approvals and all integrations.

mild.run

Ready to run
agents mild?

Join developers shipping AI agents with confidence. Free tier, no credit card required.