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.

Autonomous experiment and improvement loop v1.0.0

Use this resource to improve code iteratively against a measurable metric, through automated trials

New

  • 536 downloads
  • 167 likes
  • 139 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install autonomous-experiment-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 has you define the goal, the measurable metric, the scope and the constraints, then creates a branch and a baseline. It then runs changes, tests and measurements, keeping or discarding each trial. You get the trials kept and those discarded.

Mostly picked by

Updated November 2, 2026

Card for the “Autonomous experiment and 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: interactive setup, from goal to experiment budget
  • SKILL.md: experiment loop and agent behaviour rules

Good fit for

  • You want to speed up a process by measuring every trial
  • A function must improve on a given benchmark

Not for

    Tags

    • experimentation
    • optimisation
    • metric
    • loop

    About

    You say

    “Optimise this sort function's response time using the repo benchmark, without touching the API.”

    Card for the “Autonomous experiment and improvement loop” skill

    Skill published by a-tokyo 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 code iteratively against a measurable metric, through automated trials
    3. Review and adjust the result

    Content preview

    Files

    • LICENSE1 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • SKILL.md11.1 KB

    Origin and licence

    Listed unmodified

    Author
    a-tokyo
    Source
    github.com/a-tokyo/agent-skills/tree/99652ee6a78759a5ef655388497a2b889ad6c602/benchmarks/create-skill-autoresearch/companions/autoresearch
    Licence
    MIT

    Notice

    « autoresearch » (a-tokyo, 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.