tool-discovery

Tool Discovery for @j0kz/mcp-agents

Safety Notice

This listing is imported from skills.sh public index metadata. Review upstream SKILL.md and repository scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "tool-discovery" with this command: npx skills add j0kz/mcp-agents/j0kz-mcp-agents-tool-discovery

Tool Discovery for @j0kz/mcp-agents

Efficient discovery and exploration of 50 MCP tools across 7 categories using the orchestrator's meta-tools.

When to Use This Skill

  • First time using @j0kz tools - Get oriented with capabilities

  • Need a specific capability - Find the right tool by keyword

  • Exploring a domain - Browse tools by category

  • Before complex workflows - Discover available building blocks

  • Training new users - Introduce the ecosystem systematically

Quick Start

  1. Get Overview of All Capabilities

Tool: list_capabilities Input: {} Output: { categories: [ { name: "analysis", toolCount: 14, examples: ["review_file", "analyze_architecture"] }, { name: "generation", toolCount: 11, examples: ["generate_tests", "generate_jsdoc"] }, { name: "security", toolCount: 5, examples: ["scan_project", "scan_secrets"] }, { name: "refactoring", toolCount: 9, examples: ["extract_function", "remove_dead_code"] }, { name: "design", toolCount: 6, examples: ["design_rest_api", "design_schema"] }, { name: "documentation", toolCount: 5, examples: ["generate_readme", "generate_changelog"] }, { name: "orchestration", toolCount: 7, examples: ["run_workflow", "search_tools"] } ], totalTools: 50, hint: "Use search_tools({ category: "name" }) to explore a category" }

  1. Explore a Category

Tool: search_tools Input: { category: "security" } Output: { tools: [ { name: "scan_file", server: "security-scanner", frequency: "medium" }, { name: "scan_project", server: "security-scanner", frequency: "medium" }, { name: "scan_secrets", server: "security-scanner", frequency: "medium" }, { name: "scan_vulnerabilities", server: "security-scanner", frequency: "medium" }, { name: "generate_security_report", server: "security-scanner", frequency: "medium" } ], totalAvailable: 50 }

  1. Search by Keyword

Tool: search_tools Input: { query: "test coverage" } Output: { tools: [ { name: "generate_tests", relevance: 0.95 }, { name: "write_test_file", relevance: 0.88 }, { name: "batch_generate", relevance: 0.75 } ] }

  1. Load a Deferred Tool

Tool: load_tool Input: { toolName: "design_schema" } Output: { success: true, toolName: "design_schema", server: "db-schema", message: "Tool loaded successfully. You can now use design_schema." }

Tool Categories

Category Description Example Tools Count

analysis Code quality, architecture, metrics review_file, analyze_architecture 14

generation Create tests, docs, boilerplate generate_tests, generate_jsdoc 11

security Vulnerability scanning, secrets scan_project, scan_secrets 5

refactoring Code transformation, cleanup extract_function, remove_dead_code 9

design API and database schemas design_rest_api, design_schema 6

documentation README, CHANGELOG, API docs generate_readme, generate_changelog 5

orchestration Workflows, tool coordination run_workflow, search_tools 7

Tool Frequency Levels

The ecosystem uses frequency-based loading for efficiency:

Frequency Description Examples Loading

high Core tools, always needed review_file, generate_tests Always loaded

medium Common tools, often used scan_project, generate_readme Auto-loaded on demand

low Specialized tools, rare use design_schema, extract_function Manual load via load_tool

High-Frequency Tools (Always Available)

Tool: search_tools Input: { frequency: "high" } Output: { tools: [ { name: "review_file", server: "smart-reviewer" }, { name: "batch_review", server: "smart-reviewer" }, { name: "generate_tests", server: "test-generator" }, { name: "analyze_architecture", server: "architecture-analyzer" }, { name: "run_workflow", server: "orchestrator" }, { name: "search_tools", server: "orchestrator" }, { name: "load_tool", server: "orchestrator" }, { name: "list_capabilities", server: "orchestrator" } ] }

Discovery Patterns

Pattern 1: Domain-First Exploration

Use when: You know the problem domain but not the specific tool

  1. list_capabilities() → See all categories
  2. search_tools({ category: "domain" }) → Browse tools in category
  3. [optional] load_tool({ toolName: "..." }) → Load if needed
  4. Use the tool

Example: "I need to check for security issues"

// Step 1: Browse security category search_tools({ category: "security" }) // → Shows: scan_file, scan_project, scan_secrets, etc.

// Step 2: Use appropriate tool scan_project({ projectPath: "." })

Pattern 2: Keyword Search

Use when: You have a specific task or keyword in mind

  1. search_tools({ query: "your task" }) → Find relevant tools
  2. Review relevance scores
  3. Select best match
  4. Use the tool

Example: "I want to generate API documentation"

// Search for documentation tools search_tools({ query: "API documentation" }) // → Shows: generate_api_docs, generate_openapi, design_rest_api

// Use the most relevant generate_api_docs({ sourceDir: "./src" })

Pattern 3: Server-Focused Exploration

Use when: You know which server you need but not the specific tool

// List all tools from a server list_capabilities({ server: "security-scanner" }) // → Shows all 5 security-scanner tools with descriptions

Pattern 4: Find High-Impact Tools First

Use when: Starting a new project or workflow

// Get the essential high-frequency tools search_tools({ frequency: "high" }) // These are the core tools you'll use most often

Response Format Options

All discovery tools support response format optimization:

// Minimal - just counts list_capabilities({ response_format: "minimal" }) // → { categoryCount: 7, totalTools: 50 }

// Concise - summary without full details search_tools({ category: "security", response_format: "concise" }) // → { tools: [...names only], totalAvailable: 50 }

// Detailed - full information (default) list_capabilities({ response_format: "detailed" }) // → Complete categories with descriptions, examples, hints

Common Workflows

Workflow 1: New User Onboarding

  1. list_capabilities() → Get ecosystem overview
  2. search_tools({ frequency: "high" }) → Learn core tools
  3. run_workflow({ focus: "quality" }) → Try a pre-built workflow

Workflow 2: Find Tool for Task

  1. search_tools({ query: "your task" }) → Search by keyword
  2. Review results and relevance
  3. load_tool() if needed → Load low-frequency tool
  4. Use the tool

Workflow 3: Explore Domain

  1. list_capabilities() → See categories
  2. search_tools({ category: "chosen" }) → Explore category
  3. Read tool descriptions
  4. Select and use appropriate tools

Meta-Tools Reference

Tool Purpose Key Parameters

list_capabilities Get category overview or server tools server? , response_format?

search_tools Find tools by keyword/category/frequency query? , category? , frequency? , server? , limit?

load_tool Load a deferred tool into context toolName , server?

Tips for Efficient Discovery

  • Start broad, then narrow - Use list_capabilities first, then search_tools

  • Use relevance scores - Higher scores mean better keyword matches

  • Check frequency - High-frequency tools are always available

  • Combine filters - { category: "security", frequency: "medium" }

  • Use response_format - "minimal" for quick checks, "detailed" for exploration

Integration with Workflows

Discovery tools work seamlessly with workflows:

// Find workflow-related tools search_tools({ category: "orchestration" })

// List available workflows list_workflows()

// Run a workflow run_workflow({ workflow: "pre-commit" })

See Also

  • Code Quality Pipeline - For systematic quality improvement

  • MCP Workflow Composition - For combining tools into workflows

  • MCP Troubleshooting - When tools aren't working as expected

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

Automation

api-integration

No summary provided by upstream source.

Repository SourceNeeds Review
Automation

dependency-doctor

No summary provided by upstream source.

Repository SourceNeeds Review
Automation

brand-guidelines

No summary provided by upstream source.

Repository SourceNeeds Review
Automation

debug-detective

No summary provided by upstream source.

Repository SourceNeeds Review