MCP HubMCP Hub
返回技能列表

skill-adapter

jeremylongshore
更新于 Today
66 次查看
712
74
712
在 GitHub 上查看
general

关于

The skill-adapter analyzes existing Claude Code plugins to extract their capabilities and adapts those skills to your current task. Use it when you need functionality from other plugins without installing them directly. It acts as a universal skill chameleon, learning and applying patterns from the plugin marketplace.

快速安装

Claude Code

推荐
插件命令推荐
/plugin add https://github.com/jeremylongshore/claude-code-plugins-plus
Git 克隆备选方式
git clone https://github.com/jeremylongshore/claude-code-plugins-plus.git ~/.claude/skills/skill-adapter

在 Claude Code 中复制并粘贴此命令以安装该技能

技能文档

Skill Adapter - Universal Plugin Capability Extractor

Purpose

Analyzes plugins in the claude-code-plugins marketplace to understand their capabilities, extracts the core patterns and approaches, then adapts those skills to solve the current user's task. Acts as a "skill chameleon" that can adopt any plugin's capabilities.

How It Works

1. Task Analysis

When user presents a task:

  • Identify the core capability needed (e.g., "analyze code quality", "generate documentation", "automate deployment")
  • Determine the domain (security, devops, testing, etc.)
  • Extract key requirements and constraints

2. Plugin Discovery

Search existing plugins for relevant capabilities:

# Find plugins in relevant category
ls plugins/community/ plugins/packages/ plugins/examples/

# Search for keywords in plugin descriptions
grep -r "keyword" --include="plugin.json" plugins/

# Find similar commands/agents
grep -r "capability-name" --include="*.md" plugins/

3. Capability Extraction

For each relevant plugin found, analyze:

Commands (commands/*.md):

  • Read the markdown content
  • Extract the approach/methodology
  • Identify input/output patterns
  • Note any scripts or tools used

Agents (agents/*.md):

  • Understand the agent's role
  • Extract problem-solving approach
  • Note decision-making patterns
  • Identify expertise areas

Skills (skills/*/SKILL.md):

  • Read the skill instructions
  • Extract core capability
  • Note trigger conditions
  • Understand tool usage patterns

**Scripts (scripts/.sh, .py):

  • Analyze script logic
  • Extract reusable patterns
  • Identify best practices
  • Note error handling approaches

4. Pattern Synthesis

Combine learned patterns:

  • Merge multiple approaches if beneficial
  • Adapt to current context and constraints
  • Simplify or enhance based on user needs
  • Ensure compatibility with current environment

5. Skill Application

Apply the adapted skill:

  • Use the learned approach
  • Follow the extracted patterns
  • Apply best practices discovered
  • Adapt syntax/tools to current context

Example Workflows

Example 1: Learning Code Analysis from Security Plugins

User task: "Analyze this codebase for issues"

Process:

  1. Search for security and code-analysis plugins
  2. Find: owasp-top-10-scanner, code-quality-enforcer, security-audit-agent
  3. Extract patterns:
    • OWASP scanner checks for: SQL injection, XSS, CSRF, auth issues
    • Quality enforcer looks at: complexity, duplication, standards
    • Audit agent examines: dependencies, secrets, permissions
  4. Synthesize approach:
    • Run multi-layer analysis
    • Check security patterns first
    • Then code quality metrics
    • Then dependency issues
  5. Apply to user's codebase with adapted checks

Example 2: Adopting Documentation Skills

User task: "Generate API documentation"

Process:

  1. Find documentation plugins
  2. Discover: api-documenter, openapi-generator, readme-builder
  3. Extract approaches:
    • API documenter: parses code, generates OpenAPI spec
    • OpenAPI generator: creates interactive docs
    • README builder: structures documentation hierarchically
  4. Synthesize:
    • Parse code for endpoints
    • Generate OpenAPI/Swagger spec
    • Create interactive documentation
    • Build comprehensive README
  5. Apply combined approach to user's API

Example 3: Learning Automation from DevOps Plugins

User task: "Automate deployment process"

Process:

  1. Search DevOps category
  2. Find: deployment-automation, ci-cd-pipeline, docker-compose-generator
  3. Extract patterns:
    • Deployment automation: build → test → deploy → verify
    • CI/CD pipeline: trigger conditions, parallel jobs, rollback
    • Docker compose: service orchestration, environment management
  4. Synthesize deployment workflow
  5. Apply to user's specific tech stack

Reasoning Process

When to Use Skill Adapter

Trigger when:

  • User needs capability that might exist in marketplace
  • Task could benefit from existing plugin patterns
  • User asks: "Is there a plugin for this?"
  • Similar problems have been solved before
  • Multiple approaches could be combined

Plugin Selection Criteria

Choose plugins based on:

  1. Relevance: Matches task domain/requirements
  2. Quality: Well-documented, clear approach
  3. Simplicity: Not overly complex for the task
  4. Recency: Updated plugins preferred
  5. Popularity: Featured or well-maintained plugins

Adaptation Strategy

When adapting skills:

  • Keep: Core logic and proven patterns
  • Adapt: Syntax, tool names, specific commands
  • Enhance: Add error handling, user feedback
  • Simplify: Remove unnecessary complexity
  • Contextualize: Adjust to user's environment

Limitations and Boundaries

What Skill Adapter CAN Do:

✅ Read and analyze plugin source code ✅ Extract patterns and approaches ✅ Adapt methodologies to new contexts ✅ Combine multiple plugin capabilities ✅ Apply learned skills with reasoning

What Skill Adapter CANNOT Do:

❌ Execute compiled code (MCP servers) ❌ Access external APIs without credentials ❌ Modify original plugins ❌ Guarantee exact plugin behavior replication ❌ Use plugins that require specific environment setup

Success Criteria

Skill adaptation is successful when:

  1. User's task is completed effectively
  2. Approach borrowed makes logical sense
  3. Adapted skill is properly contextualized
  4. User understands where the approach came from
  5. Result quality matches or exceeds original plugin

Transparency

Always inform user:

  • Which plugins were analyzed
  • What patterns were extracted
  • How the skill was adapted
  • Why this approach was chosen
  • Any limitations of the adaptation

Example Usage

User: "I need to validate JSON schemas in my project"

Skill Adapter Process:
1. Searches plugins for JSON validation
2. Finds: schema-validator, json-lint-enforcer
3. Extracts: ajv library usage, error formatting patterns
4. Adapts: Uses available tools (jq, node, python)
5. Applies: Validates user's schemas with detailed errors
6. Reports: "I adapted the schema-validator approach using jq
   for validation and added custom error formatting"

Meta-Learning

Skill Adapter improves by:

  • Tracking which plugins solve which tasks best
  • Learning which patterns are most reusable
  • Noting which adaptations work well
  • Building a mental model of the marketplace
  • Understanding plugin ecosystem relationships

In essence: Skill Adapter is a meta-skill that makes the entire plugin marketplace available as a learning resource, extracting and applying capabilities on-demand to solve user tasks efficiently.

GitHub 仓库

jeremylongshore/claude-code-plugins-plus
路径: backups/skills-batch-20251204-000554/plugins/examples/pi-pathfinder/skills/pi-pathfinder
aiautomationclaude-codedevopsmarketplacemcp

相关推荐技能

algorithmic-art

该Skill使用p5.js创建包含种子随机性和交互参数探索的算法艺术,适用于生成艺术、流场或粒子系统等需求。它能自动生成算法哲学文档(.md)和对应的交互式艺术代码(.html/.js),确保作品原创性避免侵权。开发者可通过定义计算美学理念快速获得可交互的艺术实现方案。

查看技能

subagent-driven-development

开发

该Skill用于在当前会话中执行包含独立任务的实施计划,它会为每个任务分派一个全新的子代理并在任务间进行代码审查。这种"全新子代理+任务间审查"的模式既能保障代码质量,又能实现快速迭代。适合需要在当前会话中连续执行独立任务,并希望在每个任务后都有质量把关的开发场景。

查看技能

executing-plans

设计

该Skill用于当开发者提供完整实施计划时,以受控批次方式执行代码实现。它会先审阅计划并提出疑问,然后分批次执行任务(默认每批3个任务),并在批次间暂停等待审查。关键特性包括分批次执行、内置检查点和架构师审查机制,确保复杂系统实现的可控性。

查看技能

cost-optimization

其他

这个Claude Skill帮助开发者优化云成本,通过资源调整、标记策略和预留实例来降低AWS、Azure和GCP的开支。它适用于减少云支出、分析基础设施成本或实施成本治理策略的场景。关键功能包括提供成本可视化、资源规模调整指导和定价模型优化建议。

查看技能