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.

Blind skill evaluation on golden cases v1.0.0

Use this resource to evaluate and improve a skill with input and expected-output cases

New

  • 459 downloads
  • 154 likes
  • 186 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install blind-skill-evaluation-golden-cases

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 runs the skill blind, in a subagent with no context, on each example, compares the results to the expected output and uses the gaps to fix the skill. It refuses to start if one of the three required inputs is missing.

Mostly picked by

Updated November 2, 2026

Card for the “Blind skill evaluation on golden cases” skill

Details

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

Similar resources

View all

What you get

  • SKILL.md: evaluation procedure, required inputs and rules
  • SKILL.md: output format and conditions for a blind evaluation

Good fit for

  • A skill keeps producing the wrong result on a given case
  • You have input and expected-output pairs for a skill

Not for

    Tags

    • evaluation
    • skill
    • cases
    • subagent

    About

    You say

    “Evaluate this skill on my five input examples and fix the gaps with the expected output.”

    Card for the “Blind skill evaluation on golden cases” skill

    Skill published by dzhng 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 evaluate and improve a skill with input and expected-output cases
    3. Review and adjust the result

    Content preview

    Files

    • LICENSE1 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • SKILL.md7.6 KB

    Origin and licence

    Listed unmodified

    Author
    dzhng
    Source
    github.com/dzhng/jevgrep/tree/9971e04061e43f91f1d248353fd057288bfd24cf/.agents/skills/eval-skills
    Licence
    MIT

    Notice

    « eval-skills » (dzhng, 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.