Human-in-the-loop API and SDK

Human-in-the-loop for your AI agent's users.

When your agent hits a step that needs a person, Pushary asks them on their phone and waits for the answer, fail-closed.

@pushary/server · @pushary/ai-sdk · @pushary/eve · pip install pushary

What is a human-in-the-loop API?

A human-in-the-loop API lets an AI agent pause mid-run and wait for a person to decide. Pushary is a managed one: enroll(externalId) connects a user's phone in one tap, decisions.ask() blocks until they approve, deny, or answer, and the call fails closed if nobody responds. Every decision is written to a durable ledger.

New to the pattern? Human-in-the-loop for AI agents, explained covers what it means and where the checkpoints sit.

The integration

Two calls, and your agent can ask a human

Connect a user's phone once, then ask a human where it matters. Your users tap once to connect, with no account, no app, and no key on their side.

npm i @pushary/servertwo calls

import { createPusharyServer } from "@pushary/server"

const px = createPusharyServer({ apiKey: process.env.PUSHARY_API_KEY! })

// Connect an end-user's phone once. Keyless, one tap, no app to install.
await px.enroll(user.id)

// Ask a real human. Blocks until they decide. Fail-closed on timeout.
const decision = await px.decisions.ask({
  externalId: user.id,
  type: "confirm",
  question: "Approve a $480 refund to customer #4471?",
})

if (decision.approved) await issueRefund()

Every framework

Add it to the agent you already built

The same two calls, in the idiom of your framework. Published adapters where they help, and the server or Python SDK everywhere else.

Vercel AI SDK

Ask a human from your Vercel AI SDK agent, answered on a phone.

npm i @pushary/ai-sdk

LangGraph

Ask a human from your LangGraph agent, answered on a phone.

pip install pushary-langgraph

CrewAI

Ask a human from your CrewAI agent, answered on a phone.

pip install pushary-crewai

OpenAI Agents SDK

Ask a human from your OpenAI Agents SDK agent, answered on a phone.

pip install pushary-openai-agents

Eve

Ask a human from your Eve agent, answered on a phone.

npm i @pushary/eve

Mastra

Ask a human from your Mastra agent, answered on a phone.

npm i @pushary/mastra

LangChain

Ask a human from your LangChain agent, answered on a phone.

pip install pushary

MCP

Ask a human from your MCP agent, answered on a phone.

No install. One MCP server entry.

Claude Agent SDK

Ask a human from your Claude Agent SDK agent, answered on a phone.

No install. One mcpServers entry.

Hermes

Ask a human from your Hermes agent, answered on a phone.

pip install pushary

OpenClaw

Ask a human from your OpenClaw agent, answered on a phone.

npm i @pushary/server

On a framework not listed here? The same two calls work anywhere @pushary/server or pip install pushary runs.

By product type

What should not run unattended in your product

Which actions to gate, who the approver actually is, and the objection that buyer raises.

travel and expense agents

Let the agent book what fits policy on its own, and route the exception to whoever owns the budget.

customer support agents

The escalation reaches a person on their phone and comes back as an answer, not a ticket in a queue.

AI SDR and outbound agents

The rep approves the send from their lock screen, and the agent carries on without waiting for anyone to open a dashboard.

DevOps and SRE agents

Your agent gates the risky change correctly. Pushary is what gets the answer back before the incident gets worse.

legal AI agents

The supervising attorney reviews and signs off from their phone, and the record of who approved what is kept for you.

finance and accounting agents

The approver signs off from their phone, the agent blocks until they do, and the record is kept without you building one.

healthcare AI agents

The reviewing clinician approves from their phone between patients, and the review is recorded without a second system.

insurance AI agents

The adjuster or underwriter decides from their phone, inside their authority, and the record shows who it was.

security and SOC agents

The on-duty analyst authorises the containment from their phone, and the authorisation is recorded as part of the incident.

How it works

From ask to resume

Connect the phone once, ask a human at the step that guards a risky action, and act on the decision. The waiting, the delivery, and the record are handled for you.

01
Connect the phone that approves
Call enroll(externalId) with your own id for the user. They tap a one-tap link once to connect their phone. No account, no app, no key on their side.
02
Install the SDK
Use @pushary/server in TypeScript or pip install pushary in Python. On the Vercel AI SDK and Eve, use the @pushary/ai-sdk and @pushary/eve adapters.
03
Ask a human where it matters
Call decisions.ask inside the tool or step that guards the risky action. It sends the question to the person's phone and blocks until they answer.
04
Act on the fail-closed decision
If nobody answers in your window, the decision comes back denied, so a timeout never turns into an unapproved action. Every answer is written to a durable ledger.

A decision is a yes or no confirm, a multiple-choice select, or a free-text input. Every one is stored with its outcome and kept in the audit log.

FAQ

Questions, answered

What is a human-in-the-loop API?

How do I add human-in-the-loop to an AI agent I already built?

Install @pushary/server in TypeScript or pip install pushary in Python, then add two calls. enroll(externalId) connects the phone of the person who approves. decisions.ask() goes inside the tool or step that guards the risky action, and blocks until they answer. Nothing about your agent's control flow has to change, and there are published adapters for the Vercel AI SDK and Eve.

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

Human-in-the-loop for AI agents means pausing the agent at a risky step and letting a real person approve, deny, or answer before it continues. Pushary is the managed layer that does it, built so your agent can ask its own end-users. One tool call sends the question to a person's phone, blocks until they decide, fails closed if nobody answers in your window, and writes the outcome to a durable decisions ledger.

How can my agent ask my own users instead of just me?

You reference each user by your own externalId. Call enroll(externalId) to connect their phone with a one-tap link, then call decisions.ask with the same externalId. The approval lands with that specific person, so a product with thousands of users can ask each of them about their own decision.

What happens if nobody answers?

The decision fails closed. If nobody answers in the window you set, it comes back denied, and you choose what to do. Because the wait is backed by a durable ledger, a restart does not lose the request.

Do my end-users need an account or app?

No. One tap connects their phone. Delivery uses native app push when they have it, PWA push notifications otherwise, and Slack if you route it there. They never see a Pushary login or bill.

Which frameworks does Pushary support?

The Vercel AI SDK and Eve have published adapters (@pushary/ai-sdk and @pushary/eve). LangGraph, LangChain, CrewAI, Mastra, the OpenAI Agents SDK, Hermes, and OpenClaw use @pushary/server in TypeScript or pip install pushary in Python. There is also a hosted MCP server for the Claude Agent SDK and any MCP client.

How much does it cost?

The Partner plan is $99 a month for companies embedding human-in-the-loop for their end-users. There is a $9.99 a month Agent plan for a solo developer who wants their own agent to notify or ask them. Card details are collected at checkout.

Give your users the final say, on their phone.

Connect a user with one tap, ask a human with one call, and act on a fail-closed decision. Every decision is saved to a durable ledger you can audit.

Partner plan, $99 a month. Your users never see a Pushary login or bill.