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

PalateKit

Menu

Request a resource

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

0 / 280

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.

Harbor task verifiers with Reward Kit v1.0.0

Use this resource to write grading criteria for a Harbor task with Reward Kit

New

  • 395 downloads
  • 157 likes
  • 105 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install harbor-task-verifiers

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 tests/ folder of a Harbor task: programmatic Python criteria, LLM or agent judges described in TOML and score aggregation. You get a verifier that produces a reward from these criteria.

Mostly picked by

Updated November 2, 2026

Card for the “Harbor task verifiers with Reward Kit” skill

Details

Generates with
Your agent's model
Licence
Apache-2.0

Similar resources

View all

What you get

  • SKILL.md: setup and programmatic criteria examples
  • SKILL.md: LLM or agent judges and judge block options

Good fit for

  • You write a task whose free-form answer must be graded
  • You want to combine deterministic tests with an agent judgment

Not for

    Tags

    • harbor
    • reward kit
    • verifier
    • evaluation

    About

    You say

    “Write the verifier for my Harbor task: a structure check and an LLM judge for correctness.”

    Card for the “Harbor task verifiers with Reward Kit” skill

    Skill published by harbor-framework under the Apache-2.0 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 write grading criteria for a Harbor task with Reward Kit
    3. Review and adjust the result

    Content preview

    Files

    • LICENSE11.1 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • SKILL.md12.8 KB

    Origin and licence

    Listed unmodified

    Author
    harbor-framework
    Source
    github.com/harbor-framework/harbor/tree/2f424cd003ee1a94b5ed262471d5e36e4948f93f/skills/rewardkit
    Licence
    Apache-2.0

    Notice

    « rewardkit » (harbor-framework, Apache-2.0) 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.