Creates a project, reports on its experiments and session history, and deletes the project when the work behind it is finished.
4actions available
Every action runs live through MCP. Nothing to build, nothing to maintain.
Ayro sets up new experiment projects for A/B testing different response styles. Your team measures which conversational approaches work best with your customers.
Ayro checks which response variations are outperforming others. Your team doubles down on successful conversation patterns and retires underperforming ones.
Ayro retrieves conversation session records for analysis. Your team reviews customer interaction patterns to identify trends and training opportunities.
Each card is one request a support team gets, and the Humanloop actions Ayro runs to close it.
Ayro lists a project's experiments with their status, configuration and metrics, and pages through its sessions to show datapoint counts and recent inputs and outputs.
Ayro creates a Humanloop project from a name and description and returns its ID, then permanently deletes that project once the evaluation work behind it is finished.
How it works
Connect once, then just ask. There is no workflow builder to learn and nothing to maintain — Ayro reads the Humanloop actions it has and picks the ones a request needs.
Authorize the Humanloop account your team already uses — one consent screen, no API keys, no mapping tables. Ayro can only do what you granted that account, and you can disconnect it at any time.
Describe the job the way you would hand it to a teammate. Ayro maps it to the Humanloop actions that close it and chains as many as the request needs.
The agent runs the actions inside the conversation the customer is already in, so nobody copies data between tabs and your team can take over at any point.
Get started freeHumanloop · Ayro
What can you do in Humanloop without me setting anything up?
These, right away:
Great, let's start with the first one.
FAQ
Setup, permissions, and the limits of what Ayro can do inside Humanloop.
Fully. Delete Project is permanent and cannot be undone, and it removes everything linked to the project including its sessions, datapoints and evaluations. It is the one action in this toolkit worth putting a human confirmation in front of.
All 4 actions your agent can call on Humanloop, straight from the live MCP connection.
This tool creates a new project in Humanloop.
This tool allows you to delete a specific project from your Humanloop organization.
This tool retrieves an array of experiments associated with a specific project in Humanloop.
This tool retrieves a paginated list of sessions for a specific project in Humanloop.
Same connection, same setup. Pick the next one your team already uses.
Pulls printed and handwritten text out of images a customer attaches, and transcribes audio clips into text the conversation can use.
Submits enhancement jobs, tracks them by hash, returns the finished result URL, and clears stored images once they are delivered.
Renders videos from saved templates, tracks render status and remaining credits, files assets into folders, and drives live streaming avatars.
Lists 3D and AR embeds, orders modeling jobs, and reports completion percentage on the ones already in progress.
Detects the language an incoming message is written in and scores how closely two names or two addresses match before anything is acted on.
Runs pipelines singly or in bulk, stops an execution by run ID, and creates, inspects, or removes the chatbots built on top of them.

Authorize the account and your agent has all 4 actions from the first conversation.
