

HITL vs HumanLayer
HITL is an active human-in-the-loop product built for any agent: pause for a person over the API or MCP, then branch, escalate and act on the answer in a visual workflow.
TL;DR: three reasons teams pick HITL
HITL keeps a person in the loop on AI agents, from a one-tap approval to a multi-stage review that branches on every answer. Here is where it pulls ahead of HumanLayer.
An active approval product: a REST API, a hosted MCP server and webhooks, built to pause any agent for a person.
Workflows that branch on each answer, with AI steps and as many Human Review steps as you need.
Reviewers answer on iPhone, Apple Watch, Android, the web inbox or Telegram.
Built for complex decisions, not just yes or no
In HITL a workflow is a graph, not a straight line. Classify the request, ask the right person, branch on what they say and merge back. Build it on a canvas, or describe it and let the workflow assistant build it for you.
Yes: Send the refund · No: Email the customer
As many reviews as the decision needs
Put several Human Review steps in one workflow: a first check, a second sign-off, an escalation, and a different reviewer on each branch.
Branch on the answer
A Condition step routes on what the person decided or on the data. An AI Classifier sends each request down the path for its category.
Paths that meet again
Branches can merge back into one path, so a refund, an escalation and a drafted reply can all finish with the same follow-up step.
Describe it, get the workflow
The built-in workflow assistant builds the graph from a plain-language prompt. Every revision is kept, so you can restore an earlier one.
Nine step types on the canvas
- TriggerStart on a schedule, an email or a spreadsheet row.
- Human ReviewAsk a person to decide, with any of the 7 response types.
- AIRun a prompt on the data so far.
- AI ClassifierSort input into categories, each with its own path, plus Other.
- ConditionBranch yes or no on the answer or on any data.
- TransformSet variables for the steps that follow.
- DataFetch from Google Sheets, an API or a table.
- ActionCall a webhook, send an email or post to Slack.
- EndClose the run once every path is done.
What about HumanLayer?
HumanLayer's site describes structured coding workflows that move through set stages: questions, research, design, structure, plan and implement. We could not find a general workflow builder with branches. HITL works for any decision, not only code.
Feature comparison
Where HITL leads comes first. Then the features HITL and HumanLayer share, from connecting agents to pricing.
| Feature | HITL | HumanLayer |
|---|---|---|
| Nonlinear workflows (branching, multi-step review) | YesBranches, AI routing, several review steps | Not listedFixed coding stages |
| Visual workflow editor | YesCanvas with 9 step types | Not listed |
| MCP server for agents | YesHosted, with sign-in | Not listed |
| REST API | Yes | Not listedApproval SDK deprecated |
| Callback webhooks | YesPer request | Not listed |
| No-code tools (n8n, Zapier, Make) | Yesn8n, Zapier, Make | Not listed |
| Answers beyond approve or deny | Yes7 response types | Not listed |
| Timeouts with a default answer | Yes | Not listed |
| iPhone app | Yes | Not listed |
| Android app | Yes | Not listed |
| Telegram | Yes | Not listed |
| Web inbox | Yesmy.hitl.sh | YesWeb app |
| Free plan | Yes500 requests / mo | YesUp to 3 members |
"Not listed" means we could not find it on HumanLayer's public site or docs in October 2026. It may still exist. "Coming soon" channels are on HITL's roadmap and not live yet.
The products, side by side
HITL's web inbox next to a product image HumanLayer shows on its own site.

HITL web inboxfrom hitl.sh
HumanLayerfrom humanlayer.comAbout HumanLayer
HumanLayer started as an SDK for adding human approvals to agent tool calls. Its GitHub repository now says that code is pretty much all deprecated. Today HumanLayer describes itself as the multiplayer control plane for your software factory: a coding agent IDE and cloud where teams review agent research, designs and plans.
Why teams pick HITL
For the people who answer the requests and the teams who run the workflows.
A focused approval layer
HITL does one job: get a decision or an answer from a person and hand it back to the agent. It does not ask you to change how you write code.
Workflows for any decision
Refunds, content review, contract sign-off or a deploy: branch on the answer, route by category and add a second reviewer where it counts.
Answers in a tap
Requests arrive as a push on iPhone, Apple Watch and Android, or as a Telegram message, with 7 response types to choose from.
Flat monthly pricing
Free for 500 requests a month and 3 humans. Paid plans are a flat $10 or $30 a month, with 3 or 5 humans included.
Why agents work well with HITL
For the agents asking the questions. HITL gives them a clear contract: ask, wait, get a typed answer.
A hosted MCP server
Add https://mcp.hitl.sh/mcp to Claude Code, Cursor or any client that supports remote MCP. Sign in once and the agent gets tools like list_loops, create_request and get_request.
A structured answer, every time
The agent picks one of 7 response types and gets back typed data: a decision, the chosen options, a number, a rating or the edited text. No parsing a chat reply.
Timeouts that keep work moving
Every request carries a default response. If nobody answers in time, the agent gets that answer and carries on instead of waiting forever.
A setup prompt the agent follows
Copy the setup prompt from hitl.sh and paste it into your agent. It asks for your API key, reads the docs at docs.hitl.sh/llms.txt and lists your loops to prove the connection works.
Add HITL to Claude Code
claude mcp add --transport http hitl https://mcp.hitl.sh/mcpQuestions
What teams ask when they weigh HITL against HumanLayer.
- What can I use instead of HumanLayer's require_approval?
- HITL. Your agent sends a request through the REST API or the MCP server and waits for the answer. Ask for approve or deny, or for choices, text, a number or a rating.
- Is the HumanLayer SDK still maintained?
- HumanLayer's GitHub repository says the code there is pretty much all deprecated and points to a rebuild at humanlayer.com. That rebuild is a coding agent IDE rather than an approval SDK.
- Can HITL handle more than one approval step?
- Yes. A HITL workflow can hold several Human Review steps, branch on each answer with a Condition step and route by an AI Classifier. The workflow assistant can build it from a plain-language prompt.
- Does HITL work with coding agents?
- Yes. Add the MCP server to Claude Code with one command, or paste HITL's setup prompt into Claude Code, Cursor or Codex.
- How does pricing compare?
- HITL is free for 500 requests a month, then $10 or $30 a month. HumanLayer is free for up to 3 members and 200 sessions a month, then $100 per user per month.
Sources
Facts about HumanLayer come from these public pages, checked in October 2026. Products change, so check them for the latest. Spot something out of date? Email [email protected].
Ship AI workflows your team can trust
Free to start. Your first workflow takes minutes.