skills.rest
qiushiyan
qiushiyanCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

harbor-cli

Manages and evaluates Harbor container registry operations via CLI.

Updated Mar 27, 2026
Search Tags:#container registry#dataset management#evaluation management#Harbor CLI#task development
One-click install
npx skills add https://github.com/qiushiyan/harbor-eval-experiments --skill harbor-cli
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: harbor-cli
Source: https://github.com/qiushiyan/harbor-eval-experiments/tree/main/.claude/skills/harbor-cli
Command: npx skills add https://github.com/qiushiyan/harbor-eval-experiments --skill harbor-cli

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides comprehensive management and evaluation capabilities for Harbor, a container registry service. It simplifies running evaluations, managing datasets, and debugging tasks within the Harbor ecosystem.

Core Features & Use Cases

  • Harbor Evaluation Management: Execute and manage evaluations using Harbor commands like harbor run, harbor jobs, and harbor trials.
  • Dataset Management: List and download datasets for use in evaluations, including custom datasets.
  • Task and Adapter Development: Scaffold new benchmarks and review their structural integrity with harbor adapters.
  • Debugging and Analysis: Analyze failing trials and view results in a web UI with harbor tasks debug and harbor view.
  • Performance Optimization: Clean Docker images and cache with harbor cache clean.

Quick Start

Use the harbor-cli skill to run an evaluation on the dataset 'terminal-bench@2.0' using the 'claude-code' agent with the 'anthropic/claude-sonnet-4-1' model.

Frequently Asked Questions about harbor-cli

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run evaluations on a dataset using the Harbor CLI?▼

To run evaluations with Harbor CLI, use the `harbor run` command, specifying your target dataset and agent. You can execute and manage evaluations using commands like `harbor jobs` and `harbor trials`.

Can I manage and download custom datasets for Harbor evaluations?▼

Yes, Harbor CLI supports dataset management. You can list available datasets and download custom datasets directly through the command line for use in your Harbor evaluations.

How do I scaffold new benchmark adapters in Harbor?▼

You can scaffold new benchmarks and review their structural integrity using the `harbor adapters` command, which provides tools for benchmark adapter development and performance optimization.

What is the best way to debug failing evaluation trials in Harbor?▼

To debug failing trials, use the `harbor tasks debug` command to analyze the failures. You can also use `harbor view` to inspect the evaluation results in a web UI.

How do I clean Docker images and free up cache space during Harbor tasks?▼

You can clean Docker images and free up cache space by running the `harbor cache clean` command, which helps optimize performance for your container registry operations.

Related Skills

Explore More Software Engineering Skills→
lava-chenlava-chen
1

docker-deploy

Build Docker images, orchestrate multi-container apps, and publish to registries.

Community
Intermediate
koki3141koki3141
2

results-analysis

Automate statistical analysis and visualization of experimental results in Python.

Community
Advanced
ldzhhxxldzhhxx
1

docker-management

Automate Docker container lifecycle, image, and Compose operations via CLI.

Community
Intermediate
qiushiyanqiushiyan

harbor-task-creator

Automate creation and management of AI agent evaluation tasks in Harbor.

Community
Advanced