Skip to content
withHuman logo

withHuman

Intercepts agent tool calls, applies approval rules, and sends consequential ones to a human reviewer before they run.

Open source · From $20/user/moActively maintained

VoiceAgentsWorkflow automationOpen source
withhuman.ai ↗XGitHubProduct Hunt
Open withhuman.aiAlternativesSet it up for my teamIs withHuman yours? Claim the card →
At a glanceswept Oct 15, 2026

≈ 1visits/mo

Little sign of use · estimate, read Oct 15, 2026

For
Developers and teams putting human approval in front of consequential AI agent actions such as refunds or payments
Price
Open source · From $20/user/mopricing page, read Oct 21, 2026
Activity
Actively maintainedrelease 8 days ago · site updated 8 days ago
Company
Metoro Inc · Domain since 2026privacy policy · domain registered
App Store
00 ratingsiPhone & iPad · updated Oct 15, 2026
Used for
  • Hold agent refunds until a named human approves them
  • Route risky agent tool calls to on-call reviewers
  • Auto-approve low-value routine agent requests by rule
read off withhuman.ai on Oct 21, 2026
Worth knowingcomputed from what we read · each with its source
Pricingpricing page, read Oct 21, 2026

Open source · From $20/user/mo

pricing page

Your datawhat they publish, each line with its page · read Oct 21, 2026

“We never use your data to train models.”

trust center ↗

These are withHuman's own statements, read from its pages. We did not audit them. A line that is missing was not found on the pages we read; it does not mean the practice is absent.

How a program reaches itfound, not claimed · read Oct 21, 2026
  • MCP serverai.withhuman/withhumanverified in the MCP registry
  • Stateanswers, requires sign-inchecked 2026-10-21
  • Address to connectapp.withhuman.ai/api/mcp/v1
  • API docswithhuman.ai/docs/reference/overviewthe vendor's page ↗

MCP server details →

This card as JSON, for your agent →

Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.

Alternatives in Voiceby use, among tools still maintained

Figure and bar: visits/mo, estimate · year: online since

Compare withHuman with

Pick up to two.

All 10 alternatives to withHuman →

All Voice tools →

Not it? Say what you need

beta

Three tools from the Atlas for the job you describe, each with what fits and what does not.

From people who use itnot part of the reading

No notes yet. What do you use withHuman for?

Sign in to leave a note
Questions this page answers
  • What is withHuman?

    Intercepts agent tool calls, applies approval rules, and sends consequential ones to a human reviewer before they run. It is listed in the Atlas under Voice and made by Metoro Inc.

  • Who is withHuman for?

    Developers and teams putting human approval in front of consequential AI agent actions such as refunds or payments

  • How much does withHuman cost?

    withHuman is open source, per its pricing page on 2026-10-21.

  • Is withHuman free?

    Yes: withHuman is free, per its pricing page on 2026-10-21.

  • Do people use withHuman?

    Little sign of use: ≈ 1 visits/mo · estimate, 2026-10.

  • Is withHuman still maintained?

    Actively maintained: release 8 days ago · site updated 8 days ago.

  • Who owns withHuman?

    withHuman is run by Metoro Inc, per its privacy policy (2026-10-21).

  • Does withHuman train AI on your data?

    No, per its trust center (2026-10-21): “We never use your data to train models.”

  • Is withHuman safe?

    We do not audit vendors. What withHuman publishes, read 2026-10-21: SOC 2 Type II stated on its trust center; on training with your data, “We never use your data to train models.”.

  • Does withHuman have an iPhone app?

    Yes: “withHuman” on the App Store, rated 0 from 0 ratings; current version released 2026-10-15.

  • What are alternatives to withHuman?

    In Voice, by use: Axym, Pushary, Kastra, klanex, Bindfort, Retell AI.

How we read a tool

No votes, no reviews, no vendor claims. Every week a crawler reads each tool's own site and a few public registries, and the words on the card are bands over what it read.

  • Use: visits to the site (estimated), installs from npm and PyPI, presence in Chrome's usage report.
  • Activity: the newest release on GitHub, npm or PyPI; the newest dated page on the site; open roles on a public jobs board.
  • Price: the vendor's own pricing page, read with the date. A figure is printed only when it is on that page.
  • The line and the use cases are written by us from the homepage, one tool at a time, and refused when they repeat the vendor's marketing.

How AI assistants read each site — the reading vendors ask us about — is on each tool's own visibility page.

How we read every tool → · By INITE Atlas Desk