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.

Evaluator-driven improvement loop v1.0.0

Use this resource to improve a single mission with a strict evaluator and a log

New

  • 393 downloads
  • 207 likes
  • 211 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install evaluator-driven-improvement-loop

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 one mission against a defined evaluator, iterates while results do not pass, and logs each evaluation in .omc/autoresearch/. It stops at the maximum runtime or another explicit end condition. You get a log of every attempt.

Mostly picked by

Updated November 2, 2026

Card for the “Evaluator-driven improvement loop” skill

Details

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

Similar resources

View all

What you get

  • SKILL.md: evaluator contract, decision logs and maximum-runtime stop
  • SKILL.md: entry conditions, with a mission and evaluator from deep-interview

Good fit for

  • An improvement must be measured against a precise criterion
  • You want a record of every attempt to resume later

Not for

    Tags

    • improvement
    • evaluator
    • log
    • iteration

    About

    You say

    “Run the API response-time improvement with the defined evaluator, for two hours at most.”

    Card for the “Evaluator-driven improvement loop” skill

    Skill published by yeachan-heo 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 a single mission with a strict evaluator and a log
    3. Review and adjust the result

    Content preview

    Files

    • LICENSE1 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • SKILL.md3.4 KB

    Origin and licence

    Listed unmodified

    Author
    yeachan-heo
    Source
    github.com/yeachan-heo/oh-my-claudecode/tree/e10206d8789b0ac4e5214eddbfbd203c1b757dea/skills/autoresearch
    Licence
    MIT

    Notice

    « autoresearch » (yeachan-heo, 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.