### Install Dependencies and Prepare Husky

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/.agent/workflows/git-commit.md

Install project dependencies and manually run the prepare script to enable Husky git hooks if needed.

```bash
npm install
npm run prepare   # run manually if needed
```

--------------------------------

### Start HTTP Server and Inspect

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/.agent/skills/dev-toolkit/SKILL.md

Commands for running the server in HTTP mode and inspecting it. Includes options for starting in debug mode and inspecting via Stdio or HTTP.

```bash
# Start HTTP server
npm run start-http

# Start HTTP server (debug mode with port 8080)
npm run start-http:debug

# Inspect using MCP Inspector (Stdio) - Recommended for CLI
npm run inspect:stdio

# Inspect using MCP Inspector (HTTP)
npm run inspect:http
```

--------------------------------

### Install Dependencies

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/CONTRIBUTING.md

Installs the required project dependencies using npm.

```bash
npm install
```

--------------------------------

### Start HTTP Server

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Run the server in HTTP mode to enable remote access and web-based integration.

```bash
# Start HTTP server on custom port and host
npx -y awesome-copilot-mcp start-http --port 8080 --host 0.0.0.0

# Output: HTTP server listening on http://0.0.0.0:8080
# Available endpoints:
#   GET  /health       - Health check
#   GET  /metadata     - Return metadata index
#   POST /mcp          - JSON-RPC endpoint
#   GET  /mcp          - SSE streaming endpoint
#   POST /mcp/cancel   - Cancel ongoing requests
```

--------------------------------

### Initialize Commitizen

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/.agent/workflows/git-commit.md

Use Commitizen to help authors create compliant commits. Install it as a development dependency.

```bash
npx commitizen init cz-conventional-changelog --save-dev --save-exact
npx cz
```

--------------------------------

### Get Recommendations via CLI

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Retrieve content recommendations based on natural language descriptions with optional type and limit filters.

```bash
# Get recommendations for a use case
npx awesome-copilot-mcp recommend "I need help with writing Python unit tests"

# Limit results and filter by type
npx awesome-copilot-mcp recommend "code review automation" --type agents --limit 5

# Example output:
# Based on description "I need help with writing Python unit tests":
#
# 1. python-testing: Python testing specialist for pytest and unittest
#    Tags: python, testing, pytest
#
# 2. test-writer: Automated test generation agent
#    Tags: testing, automation
```

--------------------------------

### Run Awesome Copilot MCP Server as HTTP

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/README.md

Start the MCP server locally to enable remote access or for OpenAPI testing. Specify the port and host for the server.

```bash
npx -y awesome-copilot-mcp start-http --port 8080 --host 0.0.0.0
```

--------------------------------

### Run MCP Server Locally

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/.agent/skills/dev-toolkit/SKILL.md

Starts the MCP server in stdio mode for local testing with Inspector or local clients. Includes a command for debugging via stdio.

```bash
# Start server
npm start

# Debug with MCP Inspector
npm run inspect:stdio
```

--------------------------------

### Example Compliant Git Commit

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/.agent/workflows/git-commit.md

Stage files and then commit with a message following the Conventional Commits specification. This message format is used by semantic-release for versioning.

```bash
git add src/foo.ts
git commit -m "feat(cli): add --dry-run option"
```

--------------------------------

### GET /metadata

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Retrieves the current metadata index.

```APIDOC
## GET /metadata

### Description
Returns the current metadata index of the awesome-copilot repository.

### Method
GET

### Endpoint
/metadata
```

--------------------------------

### GET /health

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Performs a health check on the server.

```APIDOC
## GET /health

### Description
Checks the operational status of the server.

### Method
GET

### Endpoint
/health
```

--------------------------------

### Example Non-Compliant Git Commit

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/.agent/workflows/git-commit.md

A commit message that does not follow the Conventional Commits specification. This will be rejected by commitlint hooks.

```bash
git commit -m "fixed bug"
```

--------------------------------

### Clone Repository

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/CONTRIBUTING.md

Initializes the local development environment by cloning the repository and navigating into the directory.

```bash
git clone https://github.com/yourusername/awesome-copilot-mcp.git
cd awesome-copilot-mcp
```

--------------------------------

### Explore Content via CLI

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

List available content by type with optional tag filtering and JSON output formatting.

```bash
# List all agents
npx awesome-copilot-mcp explore agents

# List skills filtered by tag with limit
npx awesome-copilot-mcp explore skills --tags python,testing --limit 5

# Output as JSON
npx awesome-copilot-mcp explore prompts --json

# Example output:
# Available AI Agents:
#
# python-dev: Python development assistant with best practices
#   Tags: python, development
#
# code-reviewer: Automated code review agent
#   Tags: review, quality
```

--------------------------------

### Configure via environment variables

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Sets server behavior and repository configuration using shell environment variables.

```bash
# Custom metadata URL (hosted or local file path)
export ACP_METADATA_URL="https://your-cdn.com/metadata.json"
# Or local file:
export ACP_METADATA_URL="/path/to/local/metadata.json"
# Or file URI:
export ACP_METADATA_URL="file:///path/to/metadata.json"

# Custom repository configuration JSON
export ACP_REPOS_JSON='{"owner":"your-org","repo":"your-fork","branch":"custom-branch"}'

# Start server with custom config
npx awesome-copilot-mcp start

# Example: Using with local metadata for development
ACP_METADATA_URL="./metadata.json" npx awesome-copilot-mcp start-http --port 3000
```

--------------------------------

### Retrieve Content Details via CLI

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Fetch detailed information for a specific content item, including its download URL.

```bash
# Get agent details
npx awesome-copilot-mcp get agent python-dev

# Get skill details with JSON output
npx awesome-copilot-mcp get skill webapp-testing --json

# Example output:
# AI Agent: python-dev
# ──────────────────────────────────────────────────
# Python development assistant with best practices
#
# Tags: python, development
#
# ──────────────────────────────────────────────────
# Download URL:
# https://raw.githubusercontent.com/github/awesome-copilot/main/agents/python-dev.agent.md
```

--------------------------------

### Build Project

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/CONTRIBUTING.md

Compiles the project to verify that the build process is successful.

```bash
npm run build
```

--------------------------------

### Initialize and use GitHubAdapter

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Handles metadata fetching, caching, and content retrieval from a GitHub repository.

```typescript
import { GitHubAdapter, RepoConfig } from 'awesome-copilot-mcp';

// Initialize adapter with configuration
const config: RepoConfig = {
  owner: 'github',
  repo: 'awesome-copilot',
  branch: 'main',
  metadataUrl: process.env.ACP_METADATA_URL  // Optional: custom metadata URL
};

const adapter = new GitHubAdapter(config);

// Fetch the complete index
const index = await adapter.fetchIndex();
console.log(`Loaded ${index.agents.length} agents, ${index.skills.length} skills`);

// Force refresh from upstream
const freshIndex = await adapter.refresh();

// Load a skill directory with all files
const skill = await adapter.loadSkillDirectory('webapp-testing');
skill.files.forEach(file => {
  console.log(`File: ${file.path}`);
  console.log(`Content length: ${file.content.length} chars`);
});

// Fetch raw file content
const content = await adapter.fetchRawFile('agents/python-dev.agent.md');

// Clear in-memory cache
adapter.clearCache();
```

--------------------------------

### Debug Awesome Copilot MCP Server via Stdio

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/README.md

Recommended method for debugging the MCP server. Use the `--no-build` flag if you have already built the project.

```bash
npx -y awesome-copilot-mcp debug --no-build
```

--------------------------------

### Tool: load_skill_directory

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Load all files from a skill directory.

```APIDOC
## Tool: load_skill_directory

### Description
Loads all files from a skill directory at once, including SKILL.md and supporting files.

### Parameters
#### Request Body
- **name** (string) - Required - The name of the skill.

### Response
#### Success Response (200)
- **name** (string) - Skill name.
- **files** (array) - List of objects containing path and content.
```

--------------------------------

### Run Test Suite with Vitest

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/.agent/skills/dev-toolkit/SKILL.md

Executes the project's test suite using Vitest. Supports running all tests, tests with coverage reports, and watch mode for continuous testing.

```bash
# Run all tests
npm test

# Run tests with coverage
npm run test:coverage

# Watch mode
npm run test:watch
```

--------------------------------

### Configure MCP Client for Awesome Copilot

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/README.md

Add this configuration to your MCP Client (e.g., Claude Desktop or VS Code) to connect to the awesome-copilot server. This ensures you always use the latest version.

```json
{
  "mcpServers": {
    "awesome-copilot": {
      "command": "npx",
      "args": ["-y", "awesome-copilot-mcp", "start"]
    }
  }
}
```

--------------------------------

### Load Skill Directory Tool

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Fetch all files associated with a specific skill directory.

```typescript
// MCP Tool Call
{
  "name": "load_skill_directory",
  "arguments": {
    "name": "webapp-testing"
  }
}

// Response
{
  "name": "webapp-testing",
  "files": [
    {
      "path": "SKILL.md",
      "content": "# Web App Testing Skill\n\nThis skill provides..."
    },
    {
      "path": "examples/demo.js",
      "content": "// Example test setup\nconst { test } = require('@playwright/test')..."
    },
    {
      "path": "templates/test-template.md",
      "content": "## Test Template\n\n### Setup\n..."
    }
  ]
}
```

--------------------------------

### Generate and Verify Metadata

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/.agent/skills/dev-toolkit/SKILL.md

Commands for managing the awesome-copilot metadata. Supports generating full or lean metadata, verifying integrity, and checking metadata size.

```bash
# Generate full metadata
npm run generate-metadata

# Generate LEAN metadata (recommended for basic dev)
npm run generate-metadata:lean

# Verify metadata integrity
npm run generate-metadata:verify

# Check metadata size
npm run check:metadata-size
```

--------------------------------

### Search Content via CLI

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Perform keyword searches across content types with support for filtering and piping to tools like jq.

```bash
# Search all content
npx awesome-copilot-mcp search "api testing"

# Search specific type with tag filter
npx awesome-copilot-mcp search "api" --type skills --tags testing --limit 10

# JSON output for scripting
npx awesome-copilot-mcp search "python" --json | jq '.items[].name'
```

--------------------------------

### Create MCPServer with stdio transport

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Sets up an MCP server using stdio for communication with MCP clients.

```typescript
import { MCPServer, RepoConfig } from 'awesome-copilot-mcp';

const config: RepoConfig = {
  owner: 'github',
  repo: 'awesome-copilot',
  branch: 'main'
};

const server = new MCPServer(config);

// Start the server (connects via stdio)
await server.start();
// Server now handles MCP protocol messages via stdin/stdout

// Graceful shutdown
process.on('SIGINT', async () => {
  await server.stop();
  process.exit(0);
});
```

--------------------------------

### Execute MCP tools

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Provides direct access to tool implementations, including support for abort signals and progress callbacks.

```typescript
import { MCPTools, GitHubAdapter } from 'awesome-copilot-mcp';

const adapter = new GitHubAdapter({
  owner: 'github',
  repo: 'awesome-copilot'
});

const tools = new MCPTools(adapter);

// Get available tool definitions
const toolDefinitions = tools.getTools();
console.log('Available tools:', toolDefinitions.map(t => t.name));

// Execute search tool
const searchResult = await tools.handleTool('search', {
  query: 'python',
  type: 'agents',
  limit: 5
});
console.log(`Found ${searchResult.count} results`);

// Execute with abort signal and progress callback
const controller = new AbortController();
const result = await tools.handleTool('search', { query: 'testing' }, {
  signal: controller.signal,
  onProgress: (progress) => console.log('Progress:', progress),
  onPartial: (partial) => console.log('Partial result:', partial)
});

// Cancel if needed
controller.abort();
```

--------------------------------

### Run Tests

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/.agent/workflows/debug.md

Execute the test suite to ensure no regressions are introduced after a fix.

```bash
npm test
```

--------------------------------

### Tool: download

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Retrieve the download URL and metadata for a specific item.

```APIDOC
## Tool: download

### Description
Returns the direct GitHub raw URL for fetching content.

### Parameters
#### Request Body
- **name** (string) - Required - The name of the item.
- **type** (string) - Optional - The type of the item to resolve name collisions.

### Response
#### Success Response (200)
- **name** (string) - Item name.
- **type** (string) - Item type.
- **url** (string) - Direct GitHub raw URL.
- **path** (string) - File path in repository.
- **description** (string) - Item description.
```

--------------------------------

### Debug Awesome Copilot MCP Server via HTTP

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/README.md

Alternative method for debugging the MCP server using an HTTP interface.

```bash
npm run inspect:http
```

--------------------------------

### Debugging and Maintenance Commands

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/.agent/skills/dev-toolkit/SKILL.md

Helper commands for debugging and maintenance tasks. Includes direct CLI debugging, archiving reports, and cleaning build artifacts.

```bash
# Debug CLI directly
npm run debug:cli

# Archive large reports
npm run archive:reports

# Clean build artifacts
npm run clean
```

--------------------------------

### Access Resource URIs

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Use the awesome:// URI scheme to access fixed metadata or parameterized templates for agents, prompts, and other resources.

```typescript
// Fixed Resource URIs
"awesome://metadata"           // Complete metadata index
"awesome://agents/index"       // List of all agents
"awesome://prompts/index"      // List of all prompts
"awesome://instructions/index" // List of all instructions
"awesome://skills/index"       // List of all skills
"awesome://collections/index"  // List of all collections
"awesome://plugins/index"      // List of all plugins
"awesome://hooks/index"        // List of all hooks
"awesome://workflows/index"    // List of all workflows

// Template Resource URIs
"awesome://agents/{name}"              // Get specific agent by name
"awesome://prompts/{name}"             // Get specific prompt by name
"awesome://skills/{name}"              // Get specific skill by name
"awesome://search/{type}/{query}"      // Search by type and query
"awesome://tags/{tag}"                 // Browse resources by tag

// Example: Read agents index
// Request: resources/read with uri="awesome://agents/index"
// Response:
[
  {
    "name": "python-dev",
    "description": "Python development assistant",
    "tags": ["python", "development"],
    "path": "agents/python-dev.agent.md",
    "url": "https://raw.githubusercontent.com/github/awesome-copilot/main/agents/python-dev.agent.md"
  }
]

// Example: Search via resource URI
// Request: resources/read with uri="awesome://search/all/testing"
// Response:
{
  "query": "testing",
  "type": "all",
  "count": 12,
  "results": [...]
}
```

--------------------------------

### Tool: search

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Search for agents, prompts, instructions, skills, collections, plugins, hooks, and workflows.

```APIDOC
## Tool: search

### Description
Search for content by keyword with optional type and tag filtering.

### Parameters
#### Request Body
- **query** (string) - Required - The search keyword.
- **type** (string) - Optional - Filter by type (agent, prompt, instruction, skill, collection, plugin, hook, workflow, all).
- **tags** (array) - Optional - Filter by tags.
- **limit** (number) - Optional - Max results (default: 10).

### Response
#### Success Response (200)
- **query** (string) - The original query.
- **count** (number) - Number of items found.
- **items** (array) - List of matching items.
```

--------------------------------

### Download Content Tool

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Retrieve the direct GitHub raw URL and metadata for a specific item.

```typescript
// MCP Tool Call
{
  "name": "download",
  "arguments": {
    "name": "python-dev",
    "type": "agent"    // Optional: helps resolve name collisions
  }
}

// Response
{
  "name": "python-dev",
  "type": "agent",
  "url": "https://raw.githubusercontent.com/github/awesome-copilot/main/agents/python-dev.agent.md",
  "path": "agents/python-dev.agent.md",
  "description": "Python development assistant with best practices"
}
```

--------------------------------

### Lint Code

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/CONTRIBUTING.md

Runs ESLint to ensure code style compliance.

```bash
npm run lint
```

--------------------------------

### Test Compliant Commit with Husky

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/.agent/workflows/git-commit.md

Make a commit with a message that adheres to the Conventional Commits specification. This should succeed when Husky hooks are active.

```bash
git commit --allow-empty -m "feat: add example"
```

--------------------------------

### Create HttpServer with HTTP/SSE transport

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Configures an HTTP server with JSON-RPC and SSE endpoints, supporting optional TLS and authentication.

```typescript
import { HttpServer, GitHubAdapter } from 'awesome-copilot-mcp';

const adapter = new GitHubAdapter({
  owner: 'github',
  repo: 'awesome-copilot',
  branch: 'main'
});

const server = new HttpServer(adapter, {
  port: 8080,
  host: '0.0.0.0',
  authToken: process.env.API_TOKEN,           // Optional: Bearer token auth
  allowedOrigins: ['https://example.com'],    // Optional: CORS origins
  allowedHosts: ['localhost', '127.0.0.1'],   // Optional: Host validation
  rateLimit: { windowMs: 60000, max: 100 },   // Optional: Rate limiting
  tls: {                                       // Optional: HTTPS
    key: fs.readFileSync('server.key'),
    cert: fs.readFileSync('server.crt')
  }
});

await server.start();
// Server listening on https://0.0.0.0:8080

// Access the Express app for custom routes
const app = server.getApp();
app.get('/custom', (req, res) => res.json({ custom: true }));

// Graceful shutdown
await server.stop();
```

--------------------------------

### Lint Code with npm

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/.agent/skills/dev-toolkit/SKILL.md

Checks the code for style and potential errors using npm scripts. Includes commands to check linting and automatically fix fixable issues.

```bash
# Check linting
npm run lint

# Fix linting issues where possible
npm run lint:fix
```

--------------------------------

### Check Code Coverage

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/CONTRIBUTING.md

Runs the test coverage report. The project target is greater than 80% coverage.

```bash
npm run test:coverage
```

--------------------------------

### Create Feature Branch

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/CONTRIBUTING.md

Creates a new branch for feature development or bug fixes.

```bash
git checkout -b feature/my-new-feature
```

--------------------------------

### Search Content Tool

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Use the search tool to find agents, prompts, or skills by keyword, type, or tags.

```typescript
// MCP Tool Call
{
  "name": "search",
  "arguments": {
    "query": "python",
    "type": "agents",           // Optional: agent, prompt, instruction, skill, collection, plugin, hook, workflow, all
    "tags": ["development"],    // Optional: filter by tags
    "limit": 5                  // Optional: max results (default: 10)
  }
}

// Response
{
  "query": "python",
  "count": 3,
  "items": [
    {
      "name": "python-dev",
      "description": "Python development assistant with best practices",
      "type": "agent",
      "tags": ["python", "development"],
      "path": "agents/python-dev.agent.md"
    },
    {
      "name": "python-testing",
      "description": "Python testing specialist for pytest and unittest",
      "type": "skill",
      "tags": ["python", "testing"],
      "path": "skills/python-testing/SKILL.md"
    }
  ]
}
```

--------------------------------

### Run CLI in Debug Mode

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/.agent/workflows/debug.md

Use this command to run the CLI in debug mode, which can help in diagnosing issues.

```bash
npm run debug:cli
```

--------------------------------

### Plugin Definition Interface

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Defines the structure for a plugin, including version, author, license, and source repository information. Use for third-party extensions.

```typescript
interface Plugin {
  name: string;
  description: string;
  tags: string[];
  version?: string;
  author?: { name?: string; url?: string };
  homepage?: string;
  repository?: string;
  license?: string;
  source?: { source?: string; repo?: string; path?: string; url?: string };
  path: string;
  url: string;
  type: 'plugin';
}
```

--------------------------------

### Push Changes

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/CONTRIBUTING.md

Pushes the local feature branch to the remote fork.

```bash
git push origin feature/my-new-feature
```

--------------------------------

### Skill Definition Interface

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Defines the structure for a skill, including optional file paths within its directory. Use for reusable AI capabilities.

```typescript
interface Skill {
  name: string;
  description: string;
  tags: string[];
  path: string;
  files?: string[];  // Relative paths within skill directory
  url: string;
  type: 'skill';
}
```

--------------------------------

### Agent Definition Interface

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Defines the structure for an AI agent, including its name, description, tags, path, URL, and type.

```typescript
interface Agent {
  name: string;
  description: string;
  tags: string[];
  path: string;
  url: string;
  type: 'agent';
}
```

--------------------------------

### Tool: refresh_metadata

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Force refresh the metadata index from the upstream GitHub repository.

```APIDOC
## Tool: refresh_metadata

### Description
Updates the in-memory cache with the latest content from the upstream repository.

### Response
#### Success Response (200)
- **status** (string) - Success status.
- **count** (number) - Number of items indexed.
- **updated** (string) - Timestamp of the update.
```

--------------------------------

### Test Non-Compliant Commit with Husky

Source: https://github.com/rbbtsn0w/awesome-copilot-mcp/blob/main/.agent/workflows/git-commit.md

Attempt to make a commit with a message that violates the Conventional Commits specification. This should be blocked by the commit-msg hook.

```bash
git commit --allow-empty -m "bad message"
```

--------------------------------

### Refresh Metadata Tool

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Force an update of the in-memory metadata index from the upstream repository.

```typescript
// MCP Tool Call
{
  "name": "refresh_metadata",
  "arguments": {}
}

// Response
{
  "status": "success",
  "count": 247,
  "updated": "2024-01-15T10:30:00.000Z"
}
```

--------------------------------

### Complete Index Data Structure

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Defines the overall structure of the metadata index, encompassing all supported content types and update information. This is the root interface for the index.

```typescript
interface IndexData {
  agents: Agent[];
  prompts: Prompt[];
  instructions: Instruction[];
  skills: Skill[];
  collections: Collection[];
  plugins: Plugin[];
  hooks: Hook[];
  workflows: Workflow[];
  lastUpdated: string;
  version?: string;
}
```

--------------------------------

### Collection Definition Interface

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Defines the structure for a collection of items, including display options for ordering and badge visibility. Use to group related resources.

```typescript
interface Collection {
  id: string;
  name: string;
  description: string;
  tags: string[];
  items: CollectionItem[];
  display?: { ordering?: 'alpha' | 'manual'; show_badge?: boolean };
  path: string;
  url: string;
  type: 'collection';
}
```

--------------------------------

### Collection Item Interface

Source: https://context7.com/rbbtsn0w/awesome-copilot-mcp/llms.txt

Defines the structure for an item within a collection, specifying its path and kind (prompt, instruction, agent, or skill).

```typescript
interface CollectionItem {
  path: string;
  kind: 'prompt' | 'instruction' | 'agent' | 'skill';
}
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.