withHuman · Office #41,766

What it is

Calls itself “approval layer”.

• Calls itself: approval layer

“…withHuman”

“…approval layer for AI agents”

Where it puts itself

Claims “best provider”.

• Superlative claim: best provider

“We aim to be the best provider, never the only one.”

What it costs

Monthly subscription. A free tier, free forever. Trial: “3 months”. Has an enterprise plan. Sales on request. No yearly plan. Prices quoted: $0, $20 per user / month.

• Monthly: yes
• Yearly: no
• Free plan: free forever
• Price: $0 · $20 per user / month
• Trial: 3 months
• Enterprise plan: yes
• Contact sales: yes

“Free $0 hosted”

“$20 per user / month”

Who it sells to

Sells to businesses.

• Sells to: businesses

“Rules approve or deny routine calls automatically, and risky ones go to the right person on your team.”

How it runs

A API, command-line tool and mobile app for Android and iOS. Can be self-hosted. Hosted in the cloud. Offers an API. Open source. AI is the core of the product.

• Kind: api · cli · mobile
• Runs on: Android · iOS
• Runs where: cloud / hosted · self-hosted
• API: yes
• Open source: yes
• AI role: core to the product

“HTTP API”

“Install the CLI”

What it works with

Integrates with Claude Code, Codex, CrewAI and 4 more.

• Integrates with: Claude Code · Codex · CrewAI · +4

“It works across Claude Code, Codex and agents built with LangGraph, Pydantic AI and more, through an open protocol (AAP).”

“SDKs for LangGraph, CrewAI, Mastra and other agent frameworks”

What it offers as proof

Live, no waitlist. Compliance: SOC 2 Type II. Try it without signing up. Publishes docs and a blog.

• Compliance: SOC 2 Type II
• Instant signup: yes
• Trial asks for: nothing

“Sign in Get started”

“© 2026 withHuman SOC 2 Type II”