When does withHuman check an agent action?

It intercepts the agent’s tool call before the call runs.

Does every tool call need human approval?

No. Rules can automatically approve or deny routine calls, while risky calls go to a person on the team.

Which agent tools does withHuman support?

The listed tools include Claude Code, Codex, LangGraph, and Pydantic AI. withHuman uses an open protocol called AAP.

Is withHuman open source or hosted?

withHuman is open source and also offers a hosted version.

Coding / product dossier

withHuman

Intercepts AI agents’ tool calls, applies approval rules and routes risky actions to human reviewers.

Open product Daily selection

Product brief

What withHuman does.

withHuman is an approval layer for AI agents. It intercepts your agents’ tool calls before they run. Rules approve or deny routine calls automatically, and risky ones go to the right person on your team. It works across Claude Code, Codex and agents built with LangGraph, Pydantic AI and more, through an open protocol (AAP). One set of rules for every agent your company runs. Open source, with a hosted version. Use code PRODUCTHUNT for 3 months of Scale free - valid till Friday!

Superpower Daily

Your reading space

Notifications

Superpower Daily

Your reading space

withHuman

What withHuman does.

What it can help with.

Tool-call interception

Rule-based decisions

Human review routing

Shared rules across agents

Before you open it.

When we picked it.

withHuman

What withHuman does.

What it can help with.

Tool-call interception

Rule-based decisions

Human review routing

Shared rules across agents

Before you open it.

When we picked it.

Hindsight 1.0 + Memlake

CLI Manager

iwant