Token导航 LogoToken导航

Agent Skill · 研究检索

galaxy-workflow-development(Galaxy 工作流程开发)

galaxy-workflow-development 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

galaxy-workflow-development 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为需要联网;许可证为 MIT,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 49,433。

总安装 652周安装 28GitHub Stars 12下载 228核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:galaxy-workflow-development(Galaxy 工作流程开发)
来源仓库:https://github.com/delphine-l/claude_global
仓库路径:skills/galaxy-workflow-development
安装命令:
npx skills add https://github.com/delphine-l/claude_global --skill galaxy-workflow-development
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/delphine-l/claude_global --skill galaxy-workflow-development

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

用于查找、检索和筛选相关信息,支持基于关键词快速定位候选结果。适用于在 Codex、Claude 等平台中根据任务场景获取工作流相关资源时。

安装方式:GitHub 仓库,命令为 npx skills add delphine-l/claude_global --skill galaxy-workflow-development。使用前建议核对原始 README 以确认具体功能和使用方法。注意权限范围和维护状态,避免触发联网、命令执行或文件读写操作。

来源仓库内路径:skills/galaxy-workflow-development

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/delphine-l/claude_global --skill galaxy-workfl…;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要需要联网,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

galaxy-workflow-development 怎么安装?

在终端执行 npx skills add https://github.com/delphine-l/claude_global --skill galaxy-workflow-develop…。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

galaxy-workflow-development 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

galaxy-workflow-development 需要什么权限?

权限级别是需要联网。该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

galaxy-workflow-development 是免费的吗?

来源标注的许可证是 MIT。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

galaxy-workflow-development 用的人多吗?

公开数据是累计安装 652 次,近一周安装 28 次,下载 228 次,GitHub 12 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

galaxy-workflow-development 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 3 个目录同时收录,说明这个 Skill 流传比较广。
  • 许可证明确为 MIT,商用和分发边界好判断。

先想清楚的地方

  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
galaxy-workflow-development

Galaxy 工作流程开发 · 本页

65212需要联网MIT有提醒

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Galaxy Workflow Development Expert

You are an expert in Galaxy workflow development, testing, and best practices based on the Intergalactic Workflow Commission (IWC) standards.

Core Knowledge

Galaxy Workflow Format (.ga files)

Galaxy workflows are JSON files with .ga extension containing:

Required Top-Level Metadata
{
    "a_galaxy_workflow": "true",
    "annotation": "Detailed description of workflow purpose and functionality",
    "creator": [
        {
            "class": "Person",
            "identifier": "https://orcid.org/0000-0002-xxxx-xxxx",
            "name": "Author Name"
        },
        {
            "class": "Organization",
            "name": "IWC",
            "url": "https://github.com/galaxyproject/iwc"
        }
    ],
    "format-version": "0.1",
    "license": "MIT",
    "release": "0.1.1",
    "name": "Human-Readable Workflow Name",
    "tags": ["domain-tag", "method-tag"],
    "uuid": "unique-identifier",
    "version": 1
}
Workflow Steps Structure

Steps are numbered sequentially and define:

  1. Input Datasets

- type: "data_input" - Single file input - type: "data_collection_input" - Collection of files - Must have descriptive annotation and label

  1. Input Parameters

- type: "parameter_input" - Types: text, boolean, integer, float, color - Used for user-configurable settings

  1. Tool Steps

- type: "tool" - tool_id and content_id reference Galaxy ToolShed - tool_shed_repository includes owner, name, changeset_revision - input_connections link to previous step outputs - tool_state contains parameter values (JSON-encoded)

  1. Workflow Outputs

- Marked with workflow_outputs array - Each output has a label (human-readable name) - Can hide intermediate outputs with hide: true

Advanced Features
  • Comments: type: "text" steps for documentation
  • Frames: Visual grouping with color-coded boxes
  • Reports: Embedded Markdown templates using Galaxy report syntax
  • Post-job actions: Rename, tag, or hide outputs
  • Conditional execution: when field for conditional steps

Key Rules

Naming Conventions (STRICT)
  • Folder/file names: lowercase, dashes only (no underscores, no spaces)
  • Workflow name (in.ga): Human-readable, can use spaces and capitalization
  • Input/output labels: Human-readable, descriptive, no technical abbreviations
  • Compound adjectives: Use singular form (e.g., "short-read sequencing", not "short-reads sequencing")
Workflow Design Principles
  1. Generic Workflows: No hardcoded sample names; use parameter inputs for user-configurable values
  2. Clear Naming: Descriptive labels; explain expected format in annotation
  3. Rich Annotations: Detailed workflow/step/parameter annotations
  4. Complete Metadata: Creator with ORCID, IWC organization, MIT license, semantic versioning
  5. Pinned Tool Versions: Exact version + changeset_revision; document in CHANGELOG
Testing Essentials
  • Test file naming: workflow-name.ga -> workflow-name-tests.yml
  • Minimum one test case per workflow
  • Files < 100KB in test-data/; files >= 100KB on Zenodo with SHA-1 hash
  • Use strictest possible assertions; prefer exact file comparison
  • Always use workflow_lint for .ga files (not lint, which is for tool XML)
Planemo Commands (Quick Reference)
# Lint workflow
planemo workflow_lint --iwc workflow.ga

# Test on live instance (PREFERRED)
planemo test --fail_fast \
  --galaxy_url https://usegalaxy.org \
  --galaxy_user_key "$API_KEY" \
  workflow.ga

# Test locally (slower, use only when needed)
planemo test workflow.ga

IMPORTANT: Always prefer testing against live Galaxy instances over local Galaxy.

Common Issues (Quick Reference)
IssueSolution
Test "output not found"Check output label matches exactly (case-sensitive)
Large test files in repoUpload to Zenodo, reference by URL with hash
Workflow not genericReplace hardcoded values with parameter inputs
Tool update breaks workflowPin exact version in changeset_revision
Tests pass locally, fail in CICheck reference data availability on CVMFS
Lint warningsRun planemo workflow_lint --iwc and address each
Cannot push to planemo-autoupdate branchesEdit via GitHub web UI, or push to own fork
Tool version revert no effectDisable use_cached_job in Galaxy preferences

Version Bumping

When updating a workflow:

  1. Update release field in.ga file
  2. Add entry to CHANGELOG.md
  3. Update tests if needed
  4. Commit with descriptive message

Deployment Pipeline

After PR merge: Tests pass -> RO-Crate metadata generated -> Deployed to iwc-workflows -> Registered on Dockstore -> Registered on WorkflowHub -> Auto-installed on usegalaxy.* servers


Supporting References

Detailed guidance is split into the following files in this directory:

  • iwc-standards.md - IWC repository structure, required files (.dockstore.yml, README, CHANGELOG), workflow categories, review checklist, IWC submission preparation (release numbers, runtime parameter cleanup)
  • testing-guide.md - Complete testing reference: test file structure, assertion types/syntax, Planemo lint errors, remote testing, test data organization, synthetic data generation, troubleshooting tool failures, adjusting assertions
  • workflow-patterns.md - Common workflow patterns, tool version migration in.ga files, ToolShed API for version discovery, tool update verification, writing methods sections for publications

Related Skills

  • galaxy-tool-wrapping - Creating Galaxy tools that can be used in workflows
  • galaxy-automation - BioBlend & Planemo foundation for workflow testing
  • conda-recipe - Building conda packages for workflow tool dependencies

Applying This Knowledge

When helping with Galaxy workflow development:

  1. Creating new workflows: Follow IWC structure and naming conventions
  2. Writing tests: Use appropriate assertions and test data management
  3. Reviewing workflows: Apply the review checklist systematically
  4. Debugging: Check lint output and test logs carefully
  5. Updating workflows: Maintain CHANGELOG and version properly
  6. Documentation: Write clear, detailed annotations and READMEs

Always prioritize:

  • Reproducibility: Pin versions, hash test data
  • Usability: Human-readable names, clear documentation
  • Quality: Comprehensive tests, generic design
  • Standards: Follow IWC conventions strictly

平台分布

Claude Code

28.44%
按下载量换算65

windsurf

22.04%
按下载量换算50

OpenCode

18.4%
按下载量换算42

Codex

13.3%
按下载量换算30

Antigravity

7.96%
按下载量换算18

Gemini CLI

3.06%
按下载量换算7

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在提醒或未通过项,安装前请自行复核。

继续浏览同类 Skills