Can Hubi actually start an evaluation, or only read results?

Where your HoneyHive setup exposes run creation through the API, Hubi can kick off an evaluation; otherwise it focuses on reading existing runs, traces, and metrics. It confirms before launching a run since that consumes resources.

Does this work across multiple projects?

Yes, as long as your API key has access to them. You can ask Hubi to list your projects and then query traces, datasets, or evaluations within whichever one you name.

How fresh are the traces and metrics Hubi returns?

HoneyHive ingests telemetry continuously, so traces and metrics Hubi pulls reflect the platform's latest processed data rather than a fixed daily snapshot.

Will Hubi change my datasets without asking?

Reading datasets and runs happens freely, but for edits like adding or removing datapoints Hubi shows you the change first so you don't accidentally alter the data an eval depends on.

Can Hubi connect eval outcomes to my other tools?

Yes. Hubi can chain HoneyHive with anything else you've connected - for example pulling a run's score and posting it to Slack or logging it in a tracking sheet in one request.

All Integrations

Honeyhive

Artificial Intelligence Developer Tools

Shipping an LLM feature means watching traces, running evals, and managing datasets, but most of that lives behind the HoneyHive UI. Hubi queries your projects and traces, kicks off and reads evaluation runs, manages datasets, and pulls the metrics that tell you whether a prompt change actually helped - asked in plain language.

42 tools available

USE CASES

How businesses use Honeyhive + Hubi AI

Real automation scenarios powered by this integration.

📈

Check how the latest eval run scored

Ask Hubi for a project's recent evaluation and it pulls the run's metrics and pass rates into a plain summary.

Get EvaluationList RunsGet Metrics

🔬

Inspect the traces behind a bad output

Point Hubi at a session or trace and it retrieves the events and spans so you can see exactly what the model did.

Get SessionList EventsGet Trace

🗂️

Add examples to an eval dataset

Hand Hubi new test cases and it appends them to the right HoneyHive dataset so your next evaluation covers them.

Create DatapointUpdate DatasetGet Dataset

🔔

Post eval results into your team channel

After a run finishes, Hubi can summarize the scores and send them to Slack so the team sees the regression or win fast.

Get EvaluationGet MetricsSend Message

HOW IT WORKS

Get started in 3 steps

No workflows to build. Just connect, chat, and automate.

Connect Honeyhive

Authorize your account with one click. Hubi AI gets secure access to 42 tools.

Talk to your agent

Describe what you need in plain language. Your AI agent understands context and picks the right tools.

Watch it work

The agent reads, sends, labels, and organizes on your behalf - no manual setup required.

Everything you need to know about Honeyhive + Hubi AI

The Hubi AI Honeyhive integration connects your tools and data to your AI marketing agents. Start automating workflows in minutes.

Hubi connects using a HoneyHive API key from your account. Its reach matches that key's permissions across your projects, datasets, and evaluation runs.

Ready to connect Honeyhive to Hubi AI?

Start automating your workflows in minutes. No credit card required.