Back to Details

agentic-contribution-skill

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

/agentic-contribution-skill Skill

/agentic-contribution-skill Skill

Interactive AI assistant for creating production-ready skills and agentic packs for Red Hat products and platforms with automated quality validation.
All skills created follow Red Hat product guidelines, official documentation standards (docs.redhat.com, access.redhat.com), and best practices for Red Hat Enterprise Linux, OpenShift Container Platform, Ansible Automation Platform, Red Hat Lightspeed, and other Red Hat ecosystem products.
Creates: Complete skill structure with YAML frontmatter, all mandatory sections, pack integration, and new agentic packs (Lola-compatible) Validates: Tier 1 (agentskills.io) + Tier 2 (repository design principles) Applies: Red Hat documentation compliance (uses official Red Hat documentation to adapt skill content to manufacturer guidelines - not automated validation) Marketplace: Registers packs in 
marketplace/rh-agentic-collection.yml
for Lola package manager installation
一款交互式AI助手，用于为Red Hat产品及平台创建生产级Skill与Agentic Pack，并提供自动化质量验证。
所有创建的Skill均遵循Red Hat产品指南、官方文档标准（docs.redhat.com、access.redhat.com），以及Red Hat Enterprise Linux、OpenShift Container Platform、Ansible Automation Platform、Red Hat Lightspeed等Red Hat生态产品的最佳实践。
创建内容：包含YAML前置元数据、所有必填章节、Pack集成的完整Skill结构，以及兼容Lola的新Agentic Pack 验证标准：Tier 1（agentskills.io）+ Tier 2（仓库设计原则） 合规适配：Red Hat文档合规性（使用Red Hat官方文档调整Skill内容以符合厂商规范——非自动化验证） 市场注册：在
marketplace/rh-agentic-collection.yml
中注册Pack，支持Lola包管理器安装

Prerequisites

前置条件

Required Tools:
  • git
    - Version control
  • uv
    - Python environment manager
  • bash
    - Shell for validation scripts
Verification:
bash
test -d .git && echo "✓ Git repo" || echo "✗ Not a git repo"
which uv >/dev/null && echo "✓ uv installed" || echo "✗ Install uv"
test -f SKILL_DESIGN_PRINCIPLES.md && echo "✓ Valid repo" || echo "✗ Wrong directory"
Human Notification Protocol:
If prerequisites fail:
❌ Cannot execute: <issue>
📋 Setup: <specific_steps>
🔗 Doc: CONTRIBUTING.md
Security: Never display git credentials or secrets.
必备工具：
  • git
    - 版本控制工具
  • uv
    - Python环境管理器
  • bash
    - 用于验证脚本的Shell
验证命令：
bash
test -d .git && echo "✓ Git repo" || echo "✗ Not a git repo"
which uv >/dev/null && echo "✓ uv installed" || echo "✗ Install uv"
test -f SKILL_DESIGN_PRINCIPLES.md && echo "✓ Valid repo" || echo "✗ Wrong directory"
人工通知规则：
若前置条件不满足：
❌ 无法执行：<问题描述>
📋 配置步骤：<具体步骤>
🔗 文档：CONTRIBUTING.md
安全注意：绝不可显示git凭证或机密信息。

When to Use This Skill

适用场景

Use when:
  • Creating new skill for any pack
  • Importing an existing SKILL.md into the repository
  • Creating new agentic pack collection
  • Developing or editing skills for this agentic collection
  • User explicitly invokes 
    /agentic-contribution-skill
Do NOT use when:
  • Asking about standards (direct to SKILL_DESIGN_PRINCIPLES.md)
  • General contributing questions (direct to CONTRIBUTING.md)
使用时机：
  • 为任意Pack创建新Skill
  • 将现有SKILL.md导入仓库
  • 创建新的Agentic Pack集合
  • 开发或编辑本Agentic集合的Skill
  • 用户显式调用
    /agentic-contribution-skill
禁止使用时机：
  • 咨询标准相关问题（引导至SKILL_DESIGN_PRINCIPLES.md）
  • 一般性贡献问题（引导至CONTRIBUTING.md）

Workflow

工作流程

Phase 0: Mode Selection

阶段0：模式选择

Ask: "Are you creating a new skill from scratch, or importing an existing SKILL.md?"
  • Create → Proceed to Phase 1 (Discovery)
  • Import → Proceed to Phase 1-Import (Analysis)
询问："您是要从零创建新Skill，还是导入现有SKILL.md？"
  • 创建 → 进入阶段1（探索）
  • 导入 → 进入阶段1-导入（分析）

Phase 1: Discovery (5 questions max)

阶段1：探索（最多5个问题）

Ask concisely, validate before proceeding. Make additional questions if needed to gather complete context.
  1. Purpose & Red Hat Product: "What does the skill do? (1 sentence) For which Red Hat product(s) is this skill targeted?"
  2. Persona: "What role uses it?" (detect existing pack or suggest new)
  3. Pack: "Use <existing-pack>? (yes/no/create-new)"
  4. MCP Tools:
    • Read pack's 
      mcps.json
      to list existing MCP servers
    • If MCP server mentioned, verify it exists in the file
    • For tools mentioned, verify they exist in MCP server documentation or configuration
    • Ask: "What MCP tools does this skill need? Available MCPs in <pack>: <list>. Will you use existing MCPs, need new ones, or both?"
    • Never suggest tools based on intuition - only recommend tools you've verified exist
  5. Operation Type: "Read-only, additive, or destructive?" (determines color)
Color Mapping (risk-based, with Red Hat/OpenShift examples):
ColorUse CaseExamples
🔵 cyanRead-only operationslist clusters, view VM status, check node health, get CVE data
🟢 greenAdditive operationscreate VM, deploy cluster, generate playbook, add backup
🔵 blueReversible operationsrestart pod, pause MCP, start VM, stop service
🟡 yellowDestructive but recoverabledelete snapshot, remove annotation, uncordon node
🔴 redCritical/irreversibleupgrade cluster, delete cluster, restore ETCD, execute remediation
Validation:
  • Purpose: specific, under 100 chars
  • Red Hat product: identified (OpenShift, RHEL, Ansible, etc.)
  • Persona: matches known or justifies new pack
  • MCP tools: verified to exist (no assumptions/intuition)
  • Operation type → Color selected from table above
简洁提问，验证后再推进。如需收集完整上下文，可补充额外问题。
  1. 用途与Red Hat产品："该Skill的功能是什么？（一句话描述）针对哪些Red Hat产品？"
  2. 用户角色："哪些角色会使用它？"（检测现有Pack或建议新Pack）
  3. Pack选择："是否使用<现有Pack>？（是/否/创建新Pack）"
  4. MCP工具：
    • 读取Pack的
      mcps.json
      文件，列出现有MCP服务器
    • 若提及MCP服务器，验证其是否存在于文件中
    • 对于提及的工具，验证其是否存在于MCP服务器文档或配置中
    • 询问："该Skill需要哪些MCP工具？<Pack>中可用的MCP：<列表>。您将使用现有MCP、需要新MCP，还是两者都用？"
    • 绝不可凭直觉推荐工具 - 仅推荐已验证存在的工具
  5. 操作类型："只读、添加型还是破坏性操作？"（用于确定颜色标识）
颜色映射（基于风险，附Red Hat/OpenShift示例）：
颜色使用场景示例
🔵 cyan只读操作列出集群、查看VM状态、检查节点健康、获取CVE数据
🟢 green添加型操作创建VM、部署集群、生成Playbook、添加备份
🔵 blue可逆操作重启Pod、暂停MCP、启动VM、停止服务
🟡 yellow破坏性但可恢复操作删除快照、移除注解、解除节点封锁
🔴 red关键/不可逆操作升级集群、删除集群、恢复ETCD、执行修复
验证要求：
  • 用途：具体明确，不超过100字符
  • Red Hat产品：已明确（如OpenShift、RHEL、Ansible等）
  • 用户角色：匹配已知角色或有充分理由创建新Pack
  • MCP工具：已验证存在（无假设/直觉判断）
  • 操作类型 → 从上述表格中选择对应颜色

Phase 2: Definition (6 questions max)

阶段2：定义（最多6个问题）

  1. Name: Ask "Skill name? (kebab-case, unique)" → Validate if name is representative of purpose and Red Hat product. If NOT representative, propose 2-3 better alternatives and ask user to choose → Check: 
    test -d <pack>/skills/<name>/
    for uniqueness
  2. Use Cases: "3-5 user phrases for 'Use when'" (concrete, not generic)
  3. Anti-Patterns: "NOT for? (with alternative)"
  4. Workflow: "Steps with MCP tools?" (e.g., "1. Validate VM - resources_get")
  5. Common Issues: "3+ issues: problem: cause: solution"
  6. Prerequisites: "Special requirements? (env vars, permissions)"
  7. External Resources: "Any external docs/links/KB articles referenced?" (will be saved to 
    docs/
    folder)
Quality over Speed: Focus on gathering complete, accurate information. Validation and iteration will ensure correctness - prioritize quality of final result over generation time.
  1. 名称：询问"Skill名称？（短横线分隔格式，唯一）" → 验证名称是否能体现用途及对应Red Hat产品。若不具代表性，提出2-3个更优备选名称供用户选择 → 检查：
    test -d <pack>/skills/<name>/
    确认唯一性
  2. 适用场景："3-5个用户触发短语（用于'适用场景'部分）"（具体而非通用）
  3. 反模式："不适用场景？（附替代方案）"
  4. 工作流程："涉及MCP工具的步骤？"（例如："1. 验证VM - resources_get"）
  5. 常见问题："3个以上问题：问题描述：原因：解决方案"
  6. 前置条件："特殊要求？（环境变量、权限）"
  7. 外部资源："是否引用了外部文档/链接/知识库文章？"（将保存至
    docs/
    文件夹）
质量优先：专注于收集完整、准确的信息。验证和迭代将确保正确性——优先保证最终结果的质量，而非生成速度。

Phase 1-Import: Analysis (import mode only)

阶段1-导入：分析（仅导入模式）

  1. Get file: Ask for the path to the existing SKILL.md
  2. Read & parse: Read the file with Read tool. Parse YAML frontmatter, extract name, description, workflow steps, MCP tools mentioned
  3. Pack suggestion: Analyze skill content keywords and suggest the best-fit pack:
    Keywords in skill contentSuggested pack
    VM, virtual machine, KubeVirt, snapshot, migration, clonerh-virt
    CVE, vulnerability, remediation, compliance, RHEL, SRErh-sre
    deploy, build, S2I, Helm, container, route, BuildConfigrh-developer
    cluster, install, Assisted Installer, multi-cluster, ROSAocp-admin
    model, inference, GPU, vLLM, KServe, RHOAI, workbenchrh-ai-engineer
    Ansible, AAP, playbook, governance, job templaterh-automation
    CVE explanation, product lifecycle, support severity, diagnostics, patching, support case, troubleshooting, customer issue, knowledge base, must-gatherrh-basic
    • Present top match with reasoning: "This skill mentions X, Y, Z which aligns with 
      <pack>
      (persona: <role>)"
    • Ask user to confirm or override
  4. Color inference: If frontmatter has no 
    color
    , analyze the skill's workflow steps and operations to infer the risk level. Present your conclusion to the user:
    Inferred color: <color> — Reason: <operations are read-only/additive/destructive/etc.>
Confirm? (yes/override)
    Use the color mapping table from Phase 1 (Discovery).
  5. MCP tool verification: Identify all MCP tools referenced in the skill. Read the target pack's 
    mcps.json
    and verify each tool exists. Flag any tools not found — they may need a new MCP server or the skill may need adaptation.
  6. Report analysis:
    Analyzed: <path>
Name: <name> | Lines: <N> | Frontmatter: <valid/needs-fixes>
Suggested pack: <pack> (keywords: <matched>)
Color: <color> (<inferred or from frontmatter>)
MCP tools: <N verified, M not found>
Missing sections: <list or "none">

Proceed with adaptation? (yes/no/try another file)
    If user says no: ask "Would you like to try a different file, or cancel?" and act accordingly.
  1. 获取文件：询问现有SKILL.md的路径
  2. 读取与解析：使用Read工具读取文件。解析YAML前置元数据，提取名称、描述、工作流程步骤、提及的MCP工具
  3. Pack建议：分析Skill内容关键词，推荐最匹配的Pack：
    Skill内容中的关键词推荐Pack
    VM、virtual machine、KubeVirt、snapshot、migration、clonerh-virt
    CVE、vulnerability、remediation、compliance、RHEL、SRErh-sre
    deploy、build、S2I、Helm、container、route、BuildConfigrh-developer
    cluster、install、Assisted Installer、multi-cluster、ROSAocp-admin
    model、inference、GPU、vLLM、KServe、RHOAI、workbenchrh-ai-engineer
    Ansible、AAP、playbook、governance、job templaterh-automation
    CVE解释、产品生命周期、支持级别、诊断、补丁、支持案例、故障排查、客户问题、知识库、must-gatherrh-basic
    • 展示最优匹配及理由："该Skill提及X、Y、Z，与
      <pack>
      （用户角色：<角色>）匹配"
    • 询问用户确认或覆盖推荐
  4. 颜色推断：若前置元数据中无
    color
    字段，分析Skill的工作流程步骤和操作以推断风险级别。向用户展示结论：
    推断颜色：<color> — 理由：<操作属于只读/添加型/破坏性等>
是否确认？（是/覆盖）
    使用阶段1（探索）中的颜色映射表。
  5. MCP工具验证：识别Skill中引用的所有MCP工具。读取目标Pack的
    mcps.json
    文件，验证每个工具是否存在。标记未找到的工具——可能需要新增MCP服务器或调整Skill内容。
  6. 分析报告：
    已分析：<路径>
名称：<name> | 行数：<N> | 前置元数据：<有效/需修复>
推荐Pack：<pack>（匹配关键词：<匹配项>）
颜色：<color>（<推断值或来自前置元数据>）
MCP工具：<N个已验证，M个未找到>
缺失章节：<列表或“无”>

是否继续适配？（是/否/尝试其他文件）
    若用户选择否：询问"您想尝试其他文件，还是取消操作？"并按用户选择执行。

Phase 2-Import: Adaptation (import mode only)

阶段2-导入：适配（仅导入模式）

Document Consultation (REQUIRED): Read SKILL_DESIGN_PRINCIPLES.md using Read tool. Output: "I consulted SKILL_DESIGN_PRINCIPLES.md to ensure compliant adaptation."
  1. Fix frontmatter: Ensure 
    model: inherit
    and 
    color
    set (use value confirmed by user in Phase 1-Import). Add 
    metadata
    block if missing
  2. Add missing sections: Per DP6/DP7 — Prerequisites, When to Use, Workflow, Common Issues, Dependencies. Keep existing content, add structure around it
  3. Validate naming: kebab-case, check uniqueness: 
    test -d <pack>/skills/<name>/
  4. Place file: Copy to 
    <pack>/skills/<skill-name>/SKILL.md
  5. Update routing: Add entry to 
    <pack>/AGENTS.md
    intent routing table
  6. Show changes: Present summary of all modifications to user, wait for confirmation
After confirmation → proceed to Phase 5 (Validation & Iteration).
文档参考（必填）：使用Read工具读取SKILL_DESIGN_PRINCIPLES.md。输出："我已参考SKILL_DESIGN_PRINCIPLES.md以确保合规适配。"
  1. 修复前置元数据：确保
    model: inherit
    及
    color
    已设置（使用阶段1-导入中用户确认的值）。若缺失则添加
    metadata
    块
  2. 补充缺失章节：按照DP6/DP7要求——补充前置条件、适用场景、工作流程、常见问题、依赖项章节。保留现有内容，为其添加结构化框架
  3. 验证命名：短横线分隔格式，检查唯一性：
    test -d <pack>/skills/<name>/
  4. 放置文件：复制至
    <pack>/skills/<skill-name>/SKILL.md
  5. 更新路由：在
    <pack>/AGENTS.md
    意图路由表中添加条目
  6. 展示变更：向用户展示所有修改的摘要，等待确认
确认后 → 进入阶段5（验证与迭代）。

Phase 3: Pre-Generation Summary & Document Consultation

阶段3：预生成摘要与文档参考

Document Consultation (REQUIRED - Execute BEFORE generation):
  1. Action: Read SKILL_DESIGN_PRINCIPLES.md using Read tool to understand:
    • Mandatory section structure (DP7)
    • Precise workflow step format (DP2)
    • Human Notification Protocol (DP8)
    • Prerequisites template (DP8)
    • All design principles (DP1-11)
  2. Output to user: "I consulted SKILL_DESIGN_PRINCIPLES.md to ensure compliant generation."
Modularity Assessment:
  • If workflow has >10 steps OR could logically divide into phases, present modularity options to user
  • Default: Single comprehensive skill (keeps related logic together, especially for critical operations)
  • If subdivision makes sense (technical/temporal/logical separation), propose to user but let them decide
  • User decision is final
Show complete spec:
markdown
undefined
文档参考（必填 - 生成前执行）：
  1. 操作：使用Read工具读取SKILL_DESIGN_PRINCIPLES.md，了解：
    • 必填章节结构（DP7）
    • 精确的工作流程步骤格式（DP2）
    • 人工通知规则（DP8）
    • 前置条件模板（DP8）
    • 所有设计原则（DP1-11）
  2. 向用户输出："我已参考SKILL_DESIGN_PRINCIPLES.md以确保合规生成。"
模块化评估：
  • 若工作流程包含>10个步骤，或可按逻辑划分为多个阶段，向用户提供模块化选项
  • 默认选项：单一综合Skill（将相关逻辑集中在一起，尤其适用于关键操作）
  • 若拆分具有合理性（技术/时间/逻辑层面的分离），向用户提出建议但由用户最终决定
  • 用户决策为最终结果
展示完整规范：
markdown
undefined

Review Before Generation

生成前审核

Pack: <pack> | Skill: <name> | Color: <color>
Purpose: <purpose> Red Hat Product: <rh-product>
Use When: <3-5 examples> NOT for: <anti-pattern>
Workflow: <N> steps Common Issues: <N> documented MCP Tools: <tool_count> tools (verified to exist) External Resources: <count> (will be saved to docs/) Human-in-the-Loop: <Yes/No>
[If >10 steps or complex workflow]: 💡 Modularity Note: This skill has <N> steps. Options:
  1. Single comprehensive skill (recommended for critical/cohesive workflows)
  2. Subdivide into <N> modular skills (if logical separation exists)
Default: Option 1. Subdivide? (yes/no)
Proceed with generation? (yes/no)
undefined
Pack：<pack> | Skill：<name> | 颜色：<color>
用途：<purpose> Red Hat产品：<rh-product>
适用场景：<3-5个示例> 不适用场景：<反模式>
工作流程：<N>个步骤 常见问题：<N>个已记录 MCP工具：<tool_count>个工具（已验证存在） 外部资源：<count>个（将保存至docs/） 人工介入：<是/否>
[若步骤>10个或工作流程复杂]： 💡 模块化提示：该Skill包含<N>个步骤。选项：
  1. 单一综合Skill（推荐用于关键/紧密关联的工作流程）
  2. 拆分为<N>个模块化Skill（若存在逻辑分离）
默认：选项1。是否拆分？（是/否）
是否继续生成？（是/否）
undefined

Phase 4: Generation

阶段4：生成

Create structure:
bash
mkdir -p <pack>/skills/<skill-name>/
创建目录结构：
bash
mkdir -p <pack>/skills/<skill-name>/

If external resources provided by user:

若用户提供了外部资源：

mkdir -p <pack>/skills/<skill-name>/docs/

**Generate files**:
1. **SKILL.md**: YAML frontmatter + mandatory sections (follow SKILL_DESIGN_PRINCIPLES.md template - already consulted in Phase 3)
   - Focus on complete, production-ready content
   - Include all relevant information from user and verified sources
   - Keep main skill focused; detailed content can go to `docs/` if needed
2. **docs/ folder** (if applicable):
   - `docs/workflow-details.md` - Extended workflow explanations if skill is concise
   - `docs/common-issues.md` - Detailed troubleshooting with full KB article content
   - `docs/examples.md` - Comprehensive usage examples
   - `docs/external-resources.md` - Any external docs/links/KB articles mentioned by user
3. **Update <pack>/AGENTS.md**: Add intent routing entry
4. **Create <pack>/mcps.json**: If new MCP server needed (use `${ENV_VAR}` format)
5. **Update marketplace/rh-agentic-collection.yml**: If new pack (register pack for Lola installation)
6. **Create pack structure**: If new pack (README.md, AGENTS.md, skills/ directory)

Generate SKILL.md following the mandatory section template in SKILL_DESIGN_PRINCIPLES.md (already consulted in Phase 3). If SKILL.md becomes too long, move detailed content to `docs/` with references in main file.
mkdir -p <pack>/skills/<skill-name>/docs/

**生成文件**：
1. **SKILL.md**：YAML前置元数据 + 必填章节（遵循SKILL_DESIGN_PRINCIPLES.md模板——已在阶段3参考）
   - 专注于完整的生产级内容
   - 包含来自用户及已验证来源的所有相关信息
   - 保持主Skill简洁；详细内容可按需放入`docs/`
2. **docs/文件夹**（如适用）：
   - `docs/workflow-details.md` - 若Skill内容简洁，此处放置扩展的工作流程说明
   - `docs/common-issues.md` - 包含完整知识库文章内容的详细故障排查指南
   - `docs/examples.md` - 全面的使用示例
   - `docs/external-resources.md` - 用户提及的所有外部文档/链接/知识库文章
3. **更新<pack>/AGENTS.md**：添加意图路由条目
4. **创建<pack>/mcps.json**：若需要新增MCP服务器（使用`${ENV_VAR}`格式）
5. **更新marketplace/rh-agentic-collection.yml**：若创建新Pack（注册Pack以支持Lola安装）
6. **创建Pack结构**：若创建新Pack（README.md、AGENTS.md、skills/目录）

遵循SKILL_DESIGN_PRINCIPLES.md中的必填章节模板生成SKILL.md（已在阶段3参考）。若SKILL.md过长，将详细内容移至`docs/`并在主文件中添加引用。

Phase 5: Validation & Iteration

阶段5：验证与迭代

Validation always runs - quality is the priority, not speed.
Tier 1 - agentskills.io:
bash
./scripts/run-skill-linter.sh <pack>/skills/<skill-name>/
Tier 2 - Design Principles:
bash
python scripts/validate_skill_design.py <pack>/skills/<skill-name>/SKILL.md
Report clearly:
  • ✅ PASSED → Proceed to Phase 6
  • ⚠️ WARNINGS → Review warnings with user
    • Non-standard subdirectory (docs/) is acceptable if needed
    • Description buzzwords acceptable if accurate for critical skills
    • Ask: "Warnings acceptable? (yes/no)"
  • ❌ ERRORS → Fix required, iterate until validation passes
Iteration Protocol (if validation fails):
  1. Analyze errors: Identify specific issues (line count, missing sections, format problems)
  2. Determine fix strategy:
    • Line count exceeded → Move detailed content to 
      docs/
      folder, keep main skill concise
    • Missing sections → Add required sections per DP7
    • Format issues → Correct frontmatter, section headers, or structure
  3. Apply fixes: Edit SKILL.md and/or create docs/ files
  4. Re-validate: Run both Tier 1 and Tier 2 again
  5. Repeat until ✅ PASSED
Note: We iterate as many times as needed to achieve production-ready quality. Each iteration improves the skill.
验证必须执行 - 质量优先，而非速度。
Tier 1 - agentskills.io验证：
bash
./scripts/run-skill-linter.sh <pack>/skills/<skill-name>/
Tier 2 - 设计原则验证：
bash
python scripts/validate_skill_design.py <pack>/skills/<skill-name>/SKILL.md
清晰报告结果：
  • ✅ 通过 → 进入阶段6
  • ⚠️ 警告 → 与用户一起查看警告
    • 非标准子目录（docs/）若有需求则可接受
    • 若关键Skill的描述中包含流行术语且准确则可接受
    • 询问："警告是否可接受？（是/否）"
  • ❌ 错误 → 必须修复，迭代直至验证通过
迭代规则（若验证失败）：
  1. 分析错误：确定具体问题（行数超限、缺失章节、格式问题）
  2. 确定修复策略：
    • 行数超限 → 将详细内容移至
      docs/
      文件夹，保持主Skill简洁
    • 缺失章节 → 按照DP7要求补充必填章节
    • 格式问题 → 修正前置元数据、章节标题或结构
  3. 应用修复：编辑SKILL.md和/或创建docs/文件
  4. 重新验证：再次运行Tier 1和Tier 2验证
  5. 重复直至 ✅ 通过
注意：我们会进行多次迭代以达到生产级质量。每次迭代都会提升Skill的质量。

Phase 6: Post-Validation Summary

阶段6：验证后摘要

Present a concise summary to the user:
  • List all created/modified files with line counts
  • Validation results (Tier 1 + Tier 2, pass/warning counts)
  • Quality metrics (line count, section count, workflow steps, common issues)
  • Iteration count if applicable
  • Your assessment of readiness and fit within the pack
  • Ask: "Ready to commit? (yes/no)"
向用户展示简洁摘要：
  • 列出所有创建/修改的文件及行数
  • 验证结果（Tier 1 + Tier 2，通过/警告数量）
  • 质量指标（行数、章节数、工作流程步骤数、常见问题数）
  • 若有迭代，列出迭代次数
  • 您对Skill在Pack中适配性及就绪状态的评估
  • 询问："是否准备提交？（是/否）"

Phase 7: Git Workflow (Optional - User Controls)

阶段7：Git工作流程（可选 - 用户控制）

User has full control. Each step requires explicit confirmation:
  1. Branch: "Create 
    feat/<skill-name>
    ? (yes/no)"
  2. Stage: Show files, ask confirmation
  3. Commit: Propose message (
    feat: add <skill-name> skill to <pack>
    ), wait for approval
  4. Push: "Push changes? (yes/no)"
  5. PR: Use 
    gh pr create
    if available, or provide manual steps
User can skip any step.
用户拥有完全控制权。每个步骤都需要明确确认：
  1. 分支："是否创建
    feat/<skill-name>
    分支？（是/否）"
  2. 暂存：展示文件，询问确认
  3. 提交：建议提交信息（
    feat: add <skill-name> skill to <pack>
    ），等待批准
  4. 推送："是否推送变更？（是/否）"
  5. PR：若可用则使用
    gh pr create
    ，或提供手动步骤
用户可跳过任意步骤。

Phase 8: Final Summary

阶段8：最终摘要

Report: skill path, quality status, PR URL (if created), and note that CI checks will run automatically.
报告：Skill路径、质量状态、PR链接（若已创建），并说明CI检查将自动运行。

Common Issues

常见问题

Issue 1: "Description exceeds 500 tokens"

问题1："描述超过500个token"

Fix: Shorten frontmatter - move details to body sections.
修复方案：缩短前置元数据 - 将详细内容移至正文章节。

Issue 2: "Skill name exists"

问题2："Skill名称已存在"

Fix: Choose more specific name. Check: 
ls <pack>/skills/
修复方案：选择更具体的名称。检查：
ls <pack>/skills/

Issue 3: "MCP server not configured"

问题3："MCP服务器未配置"

Fix: Add to 
<pack>/mcps.json
using 
${ENV_VAR}
format.
修复方案：使用
${ENV_VAR}
格式添加至
<pack>/mcps.json
。

Issue 4: "Git push authentication failed"

问题4："Git推送认证失败"

Fix: Configure credentials:
bash
undefined
修复方案：配置凭证：
bash
undefined

HTTPS

HTTPS

git config --global credential.helper store
git config --global credential.helper store

SSH

SSH

ssh-add ~/.ssh/id_ed25519
undefined
ssh-add ~/.ssh/id_ed25519
undefined

Issue 5: "uv not found"

问题5："未找到uv"

Fix: Install:
bash
curl -LsSf https://astral.sh/uv/install.sh | sh
修复方案：安装uv：
bash
curl -LsSf https://astral.sh/uv/install.sh | sh

Issue 6: "New pack not installable via Lola"

问题6："新Pack无法通过Lola安装"

Cause: Pack not registered in 
marketplace/rh-agentic-collection.yml
Fix: Add pack entry to marketplace file:
yaml
- name: <pack-name>
  version: 0.1.0
  description: <pack-description>
  path: <pack-name>
原因：Pack未在
marketplace/rh-agentic-collection.yml
中注册
修复方案：在市场文件中添加Pack条目：
yaml
- name: <pack-name>
  version: 0.1.0
  description: <pack-description>
  path: <pack-name>

Issue 7: "Line count exceeds 500"

问题7："行数超过500"

Cause: Skill content is comprehensive but exceeds agentskills.io 500-line limit
Fix: Iterate to move detailed content to 
docs/
folder:
  1. Create 
    <skill>/docs/
    directory
  2. Move detailed workflow explanations to 
    docs/workflow-details.md
  3. Move full troubleshooting KB articles to 
    docs/common-issues.md
  4. Move comprehensive examples to 
    docs/examples.md
  5. Keep main SKILL.md concise with references to docs/
  6. Re-run validation
Example:
markdown
undefined
原因：Skill内容全面但超出agentskills.io的500行限制
修复方案：迭代将详细内容移至
docs/
文件夹：
  1. 创建
    <skill>/docs/
    目录
  2. 将详细工作流程说明移至
    docs/workflow-details.md
  3. 将完整故障排查知识库文章移至
    docs/common-issues.md
  4. 将全面示例移至
    docs/examples.md
  5. 保持主SKILL.md简洁，并添加指向docs/的引用
  6. 重新运行验证
示例：
markdown
undefined

Common Issues

常见问题

See docs/common-issues.md for detailed solutions.
详细解决方案请参阅docs/common-issues.md。

Issue 1: Snapshot Fails

问题1：快照失败

Storage doesn't support snapshots. Solution: Use snapshot-capable storage. Details
undefined
存储不支持快照。 解决方案：使用支持快照的存储。详细信息
undefined

Dependencies

依赖项

Required MCP Servers

必备MCP服务器

None - This skill operates on repository files and git operations only. No MCP servers required.
无 - 该Skill仅操作仓库文件及git操作。无需MCP服务器。

Required MCP Tools

必备MCP工具

None - Uses Claude Code built-in tools (Read, Write, Edit, Bash, Skill) for file operations and validation.
无 - 使用Claude Code内置工具（Read、Write、Edit、Bash、Skill）进行文件操作和验证。

Related Skills

相关Skill

None - agentic-contribution-skill is self-contained and doesn't invoke other skills.
无 - agentic-contribution-skill是独立的，不会调用其他Skill。

Repository Files

仓库文件

  • SKILL_DESIGN_PRINCIPLES.md
    - Design principles (DP1-11)
  • scripts/run-skill-linter.sh
    - Tier 1 validation
  • scripts/validate_skill_design.py
    - Tier 2 validation
  • Makefile
    - Validation targets
  • marketplace/rh-agentic-collection.yml
    - Lola marketplace registry
  • SKILL_DESIGN_PRINCIPLES.md
    - 设计原则（DP1-11）
  • scripts/run-skill-linter.sh
    - Tier 1验证脚本
  • scripts/validate_skill_design.py
    - Tier 2验证脚本
  • Makefile
    - 验证目标
  • marketplace/rh-agentic-collection.yml
    - Lola市场注册表

System Requirements

系统要求

See Prerequisites section for required system tools (git, uv, bash).
系统必备工具（git、uv、bash）请参阅前置条件章节。

Reference Documentation

参考文档

Internal:
  • SKILL_DESIGN_PRINCIPLES.md - Complete design principles (DP1-11)
  • CONTRIBUTING.md - Contribution workflow guide
External:
内部文档：
  • SKILL_DESIGN_PRINCIPLES.md - 完整设计原则（DP1-11）
  • CONTRIBUTING.md - 贡献工作流程指南
外部文档：

Critical: Human-in-the-Loop Requirements

关键：人工介入要求

MUST confirm before:
  1. File Generation: Show pre-generation summary, wait for "yes"
  2. Git Branch: Ask before creating branch
  3. Git Commit: Show commit message, wait for approval
  4. Git Push: Confirm before pushing to remote
  5. Quality Enforcement: Block on validation errors, warn on warnings
NEVER:
  • Create commits without confirmation
  • Push without explicit approval
  • Skip validation steps
  • Proceed if Tier 1 or Tier 2 validation fails with errors
Why: User controls their repository. Quality standards ensure CI success.
必须在以下操作前确认：
  1. 文件生成：展示预生成摘要，等待用户回复"是"
  2. Git分支：创建分支前询问用户
  3. Git提交：展示提交信息，等待用户批准
  4. Git推送：推送至远程仓库前确认
  5. 质量强制执行：验证错误会阻止流程，警告会触发提醒
绝不可：
  • 未经确认创建提交
  • 未经明确批准推送
  • 跳过验证步骤
  • 若Tier 1或Tier 2验证出现错误仍继续流程
原因：用户控制其仓库。质量标准确保CI检查成功。

Security Considerations

安全考量

  • Git Credentials: Never display tokens, SSH keys, or passwords
  • Environment Variables: Report presence only, never values
  • User Repository: Only operate in user's workspace
  • File Overwrite: Warn before overwriting, require confirmation
  • MCP Credentials: Only 
    ${ENV_VAR}
    format, never hardcoded
  • Script Execution: User's repo context only
  • Git凭证：绝不可显示令牌、SSH密钥或密码
  • 环境变量：仅报告存在性，绝不显示值
  • 用户仓库：仅在用户工作区操作
  • 文件覆盖：覆盖前发出警告，需用户确认
  • MCP凭证：仅使用
    ${ENV_VAR}
    格式，绝不硬编码
  • 脚本执行：仅在用户仓库上下文执行

Example Usage

示例用法

See docs/examples.md for comprehensive examples.
完整示例请参阅docs/examples.md。

Quick Example: Creating VM Backup Skill

快速示例：创建VM备份Skill

User: "Create skill for rh-virt to backup VMs"

Skill guides through:
1. Discovery (5 questions) - Verifies MCP tools exist
2. Definition (6 questions) - Gathers workflow details
3. Pre-generation summary - Consults SKILL_DESIGN_PRINCIPLES.md
4. Generation - Creates SKILL.md with all mandatory sections
5. Validation - Tier 1 + Tier 2, iterates if needed
6. Git workflow - User confirms each step

Result: Production-ready skill in rh-virt/skills/vm-backup-create/
More examples: docs/examples.md
  • Example 1: VM backup skill (complete create interaction)
  • Example 2: Non-representative name correction
  • Example 3: Importing an existing skill (analysis, adaptation, validation)
  • Example 4: Large skill requiring docs/ folder with iteration
用户："为rh-virt创建备份VM的Skill"

Skill引导完成：
1. 探索（5个问题）- 验证MCP工具存在
2. 定义（6个问题）- 收集工作流程细节
3. 预生成摘要 - 参考SKILL_DESIGN_PRINCIPLES.md
4. 生成 - 创建包含所有必填章节的SKILL.md
5. 验证 - Tier 1 + Tier 2，必要时迭代
6. Git工作流程 - 用户确认每个步骤

结果：生产级Skill位于rh-virt/skills/vm-backup-create/
更多示例：docs/examples.md
  • 示例1：VM备份Skill（完整创建交互）
  • 示例2：非代表性名称修正
  • 示例3：导入现有Skill（分析、适配、验证）
  • 示例4：需拆分至docs/文件夹的大型Skill及迭代过程