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.

Improving prompts from traces v1.0.0

Use this resource to improve or debug a prompt using production traces, evaluations and annotations

New

  • 586 downloads
  • 115 likes
  • 218 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install prompt-optimization-from-traces

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 extracts the prompt in use from LLM spans, gathers performance signals (evaluations, annotations, performance columns) and runs an optimisation loop based on that data. It proposes an improved version to compare with the current one. You get a prompt improved from evidence rather than guesswork.

Mostly picked by

Updated November 2, 2026

Card for the “Improving prompts from traces” skill

Details

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

Similar resources

View all

What you get

  • SKILL.md: finding prompts in trace data and signal columns
  • SKILL.md: extracting the current prompt and filtering by model
  • references/optimization-meta-prompt.md: meta-prompt used to propose a version

Good fit for

  • Your agent answers unevenly and you want a more reliable prompt
  • Your evaluations show a quality drop on one type of question

Not for

    Tags

    • optimisation
    • prompt
    • traces
    • evaluations
    • arize

    About

    You say

    “Improve my assistant's prompt using the traces where the faithfulness score is low.”

    Card for the “Improving prompts from traces” skill

    Skill published by arize-ai 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 improve or debug a prompt using production traces, evaluations and annotations
    3. Review and adjust the result

    Content preview

    Files

    • LICENSE1 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • references/ax-profiles.md5.7 KB
    • references/ax-setup.md1.7 KB
    • references/optimization-meta-prompt.md2.5 KB
    • SKILL.md18.2 KB

    Origin and licence

    Listed unmodified

    Author
    arize-ai
    Source
    github.com/arize-ai/arize-skills/tree/877c9daf7187543f6a4cecdfc24cd33c43fa3e48/skills/arize-prompt-optimization
    Licence
    MIT

    Notice

    « arize-prompt-optimization » (arize-ai, 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.