Every premium and members-only resource for $9.99/month, as a no-commitment subscription. See the plans

PalateKit

Sign in
Menu

Request a resource

Describe the resource you are missing in one sentence. We will email you when it is published.

0 / 280

A page, a video or an image online.

Unlock PalateKit

Unlock everything

Plan

Paying means accepting our terms of sale.

Secure payment by Stripe. Receipts and invoices sent by email.

Welcome back

Sign in to install, download or save this resource. A free account is enough for free resources.

Set up an online experiment on an LLM v1.0.0

Use this resource to run an A/B test in production on an LLM application

New

  • 570 downloads
  • 191 likes
  • 100 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install set-up-online-llm-experiment

With no option, the skill goes to ~/.agents/skills/; add --agent claude-code (or codex, cursor…) to target an agent.

First time: connect the CLI with

npx --yes palatekit@latest login

The agent creates the feature flag, wires the application, creates the experiment and attaches the score, token and cost metrics. It then sets the traffic split between control and treatment, and starts the experiment.

Mostly picked by

Updated November 2, 2026

Card for the “Set up an online experiment on an LLM” skill

Details

Works in
Install guides
Generates with
Your agent's model
Licence
MIT

Similar resources

View all

What you get

  • SKILL.md: discovery, the flag and wiring the application
  • SKILL.md: creating the experiment and the score, token and cost metrics
  • SKILL.md: traffic split and starting the run

Good fit for

  • You want to compare two prompts on live traffic
  • A model change must be tested before it is rolled out

Not for

    Tags

    • experiment
    • A/B test
    • LLM
    • feature flag

    About

    You say

    “Test the new prompt of our assistant against the current one on live traffic.”

    Card for the “Set up an online experiment on an LLM” skill

    Skill published by datadog-labs under the MIT license, delivered in its original version.

    Show more

    How to use it

    1. Install the skill in your agent's skills folder
    2. Ask your agent to run an A/B test in production on an LLM application
    3. Review and adjust the result

    Content preview

    Files

    • LICENSE1.1 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • SKILL.md24 KB

    Origin and licence

    Listed unmodified

    Author
    datadog-labs
    Source
    github.com/datadog-labs/agent-skills/tree/ef022309cc6b6296e7574589ac332699b1c9a092/agent-observability/agent-observability-online-experiment
    Licence
    MIT

    Notice

    « agent-observability-online-experiment » (datadog-labs, MIT) livré dans sa version d'origine : fiche en français et en anglais par PalateKit.

    User reviews

    –

    No ratings yet

    Loading reviews…

    Delete your review?

    Your rating and comment will be removed from the page.