### MCP Host Installation Guides Source: https://github.com/github/github-mcp-server Links to installation guides for various MCP (Multi-Copilot Platform) hosts, detailing how to set up AI LLM integration in different development environments. ```markdown ### Install in other MCP hosts * **[GitHub Copilot in other IDEs](/github/github-mcp-server/blob/main/docs/installation-guides/install-other-copilot-ides.md)** - Installation for JetBrains, Visual Studio, Eclipse, and Xcode with GitHub Copilot * **[Claude Applications](/github/github-mcp-server/blob/main/docs/installation-guides/install-claude.md)** - Installation guide for Claude Web, Claude Desktop and Claude Code CLI * **[Cursor](/github/github-mcp-server/blob/main/docs/installation-guides/install-cursor.md)** - Installation guide for Cursor IDE * **[Windsurf](/github/github-mcp-server/blob/main/docs/installation-guides/install-windsurf.md)** - Installation guide for Windsurf IDE ``` -------------------------------- ### React Component: Quickstart Banner Source: https://docs.perplexity.ai/ This React component, `QuickstartBanner`, renders a prominent banner designed to encourage users to start building. It features an icon, a 'Quickstart Guide' title, a descriptive subtitle, and a 'Start Building' call-to-action button that links to a quickstart page. The component uses `_jsx` and `_jsxs` for rendering and applies various Tailwind CSS-like classes for styling. ```JavaScript const QuickstartBanner = () => { const _components = { a: "a", div: "div", path: "path", span: "span", svg: "svg", ..._provideComponents() }; return _jsxs(_components.div, { className: "mint-flex mint-items-center mint-justify-between bg-card mint-border border-border mint-rounded-xl mint-p-5 md:mint-p-6 mint-w-full mint-mt-4 mint-mb-8 dark:mint-bg-[#2C3A2C]", children: [_jsxs(_components.div, { className: "mint-flex mint-items-center mint-gap-4", children: [_jsx(_components.svg, { xmlns: "http://www.w3.org/2000/svg", width: "24", height: "24", viewBox: "0 0 40 35", fill: "none", children: _jsx(_components.path, { d: "M7.3187 33C4.77908 30.4604 3.04959 27.2246 2.34892 23.7021C1.64825 20.1795 2.00788 16.5283 3.38233 13.2101C4.75678 9.89188 7.08432 7.05577 10.0706 5.0604C13.0569 3.06503 16.5678 2 20.1594 2C23.751 2 27.2619 3.06503 30.2482 5.0604C33.2345 7.05577 35.5621 9.89188 36.9365 13.2101C38.311 16.5283 38.6706 20.1795 37.9699 23.7021C37.2693 27.2246 35.5398 30.4604 33.0002 33M28.2303 12.0886L20.1594 20.1594", stroke: "currentColor", strokeWidth: "3", strokeLinecap: "round", strokeLinejoin: "round", className: "text-muted-foreground" }) }), _jsxs(_components.div, { className: "mint-flex mint-flex-col", children: [_jsx(_components.span, { className: "mint-font-semibold mint-text-base md:mint-text-lg text-foreground", children: "Quickstart Guide" }), _jsx(_components.span, { className: "mint-text-sm md:mint-text-base text-muted-foreground mint-mt-0.5", children: "Getting started is simple and fast—make your first API call within minutes." })] })] }), _jsx(_components.a, { href: "/getting-started/quickstart", className: "mint-ml-4 mint-px-4 mint-py-2 bg-primary mint-text-white mint-rounded mint-font-mono mint-text-sm mint-font-semibold mint-shadow hover:bg-primary/90 mint-transition mint-whitespace-nowrap", children: "Start Building" })] }); }; ``` -------------------------------- ### Quickstart Banner Component Source: https://docs.perplexity.ai/ A React component designed to display a quickstart guide with a call to action. It includes a title, descriptive text, and a button to initiate the building process. ```jsx function QuickstartBanner({ _components, SonarLogo, _jsxs, _jsx }) { return _jsxs(_components.div, { className: "mint-flex mint-flex-col md:mint-flex-row mint-justify-between mint-items-start md:mint-items-center mint-mt-8 mint-mb-12", children: [ _jsxs("div", { className: "mint-flex mint-items-center mint-gap-2", children: [ _jsx(SonarLogo, {}), _jsx("h1", { className: "mint-text-2xl mint-font-bold text-foreground", children: _jsx("span", { className: "mint-font-thin", children: "Developer Platform" }) }) ] }), _jsx("div", { className: "mint-mt-4 md:mint-mt-0 md:mint-ml-8 mint-max-w-md mint-text-right", children: _jsx("span", { className: "mint-text-[12px] text-muted-foreground mint-font-mono", children: _jsx(_components.p, { children: "Power your products with unparalleled real-time, web-wide research and Q&A capabilities." }) }) }) ] }); } ``` ```jsx function QuickstartBannerCTA({ _components, _jsxs, _jsx }) { return _jsxs(_components.div, { className: "mint-flex mint-items-center mint-gap-4", children: [ _jsx(_components.svg, { xmlns: "http://www.w3.org/2000/svg", width: "24", height: "24", viewBox: "0 0 40 35", fill: "none", children: _jsx(_components.path, { d: "M7.3187 33C4.77908 30.4604 3.04959 27.2246 2.34892 23.7021C1.64825 20.1795 2.00788 16.5283 3.38233 13.2101C4.75678 9.89188 7.08432 7.05577 10.0706 5.0604C13.0569 3.06503 16.5678 2 20.1594 2C23.751 2 27.2619 3.06503 30.2482 5.0604C33.2345 7.05577 35.5621 9.89188 36.9365 13.2101C38.311 16.5283 38.6706 20.1795 37.9699 23.7021C37.2693 27.2246 35.5398 30.4604 33.0002 33M28.2303 12.0886L20.1594 20.1594", stroke: "currentColor", strokeWidth: "3", strokeLinecap: "round", strokeLinejoin: "round", className: "text-muted-foreground" }) }), _jsxs("div", { className: "mint-flex mint-flex-col", children: [ _jsx("span", { className: "mint-font-semibold mint-text-base md:mint-text-lg text-foreground", children: "Quickstart Guide" }), _jsx("span", { className: "mint-text-sm md:mint-text-base text-muted-foreground mint-mt-0.5", children: "Getting started is simple and fast—make your first API call within minutes." }) ] }) ] }); } function QuickstartBannerLink({ _components }) { return _jsx(_components.a, { href: "/getting-started/quickstart", className: "mint-ml-4 mint-px-4 mint-py-2 bg-primary mint-text-white mint-rounded mint-font-mono mint-text-sm mint-font-semibold mint-shadow hover:bg-primary/90 mint-transition mint-whitespace-nowrap", children: "Start Building" }); } ``` -------------------------------- ### GitHub MCP Server Configuration Source: https://github.com/github/github-mcp-server Example JSON configuration for the GitHub MCP Server, specifying the command, arguments, and environment variables for integration. This setup allows the server to be run directly using Go build output. ```json { "mcp": { "servers": { "github": { "command": "/path/to/github-mcp-server", "args": [ "stdio" ], "env": { "GITHUB_PERSONAL_ACCESS_TOKEN": "" } } } } } ``` -------------------------------- ### Docker Setup and GitHub PAT Requirements Source: https://github.com/github/github-mcp-server To run the server in a container, Docker must be installed and running. You will also need to create a GitHub Personal Access Token (PAT) with appropriate permissions to interact with GitHub APIs. ```markdown To run the server in a container, you will need to have Docker installed. Once Docker is installed, you will also need to ensure Docker is running. The image is public; if you get errors on pull, you may have an expired token and need to `docker logout ghcr.io`. Lastly you will need to Create a GitHub Personal Access Token. The MCP server can use many of the GitHub APIs, so enable the permissions that you feel comfortable granting your AI tools (to learn more about access tokens, please check out the documentation). ``` -------------------------------- ### Install Remote GitHub MCP Server in VS Code Source: https://github.com/github/github-mcp-server Provides a direct link to install the remote GitHub MCP Server in VS Code. This installation method uses a remote server hosted by GitHub for ease of setup. It includes the necessary configuration parameters for the MCP server. ```APIDOC VSCode MCP Server Installation: URL: https://insiders.vscode.dev/redirect/mcp/install Parameters: name: github config: JSON string specifying server type and URL. - type: "http" - url: "https://api.githubcopilot.com/mcp/" Example Configuration: config=%7B%22type%22%3A%20%22http%22%2C%22url%22%3A%20%22https%3A%2F%2Fapi.githubcopilot.com%2Fmcp%2F%22%7D ``` ```JSON { "type": "http", "url": "https://api.githubcopilot.com/mcp/" } ``` -------------------------------- ### Connection URI Examples Source: https://www.postgresql.org/docs/current/libpq-connect.html Illustrative examples of valid PostgreSQL connection URIs, demonstrating various combinations of user, password, host, port, database name, and connection parameters. ```APIDOC postgresql:// postgresql://localhost postgresql://localhost:5433 postgresql://localhost/mydb postgresql://user@localhost postgresql://user:secret@localhost postgresql://other@localhost/otherdb?connect_timeout=10&application_name=myapp postgresql://host1:123,host2:456/somedb?target_session_attrs=any&application_name=myapp ``` -------------------------------- ### Web Page Node URL Configuration Example Source: https://context7_llms Example of how to configure the Web Page Node by listing URLs, one per line, for content fetching. ```example https://example.com/ https://docs.giselles.ai/glossary/node ``` -------------------------------- ### Code Examples for Use Case Source: https://docs.perplexity.ai/ Provides a set of code examples demonstrating different functionalities related to a specific use case. Includes examples for command-line execution (Bash), client-side scripting (JavaScript), and server-side logic (Python). ```bash curl ``` ```javascript javascript ``` ```python python ``` -------------------------------- ### Install Local GitHub MCP Server with VS Code Source: https://github.com/github/github-mcp-server Provides instructions and links to install the GitHub MCP Server locally using VS Code. It includes options for both stable and Insiders versions of VS Code, requiring a GitHub Personal Access Token for authentication. ```docker docker run -i --rm -e GITHUB_PERSONAL_ACCESS_TOKEN ghcr.io/github/github-mcp-server ``` -------------------------------- ### GitHub Copilot Configuration Example Source: https://github.com/github/github-mcp-server Example configuration structure for a GitHub Copilot Personal Access Token (PAT) or similar authentication credential. This snippet illustrates the expected keys and value types for setting up remote access. ```json { "promptString": "", "id": "github_mcp_pat", "description": "GitHub Personal Access Token", "password": true } ``` -------------------------------- ### Keyword/Value Connection String Example Source: https://www.postgresql.org/docs/current/libpq-connect.html An example of a connection string using the keyword/value format. Parameters are separated by spaces, and values containing spaces or special characters should be enclosed in single quotes. ```text host=localhost port=5432 dbname=mydb connect_timeout=10 ``` -------------------------------- ### Example Prompt: Calculate Number of Users Source: https://docs.giselles.ai/glossary/postgresql-tools An example prompt demonstrating how to instruct the AI model to use enabled PostgreSQL tools to perform a task. ```markdown Example Prompt: Calculate the number of users Imagine you have enabled the `getTableStructure` and `query` tools for a node. Prompt: "How many users are in the 'users' table?" ``` -------------------------------- ### Giselle AI Model Selection Guide Source: https://docs.giselles.ai/models/providers/openai Comprehensive guide for selecting the most suitable OpenAI model within Giselle. It outlines recommendations based on reasoning capabilities, speed, cost-effectiveness, context window size (up to 1M tokens), and specific task requirements like web search and multimodal needs. ```APIDOC Model Selection Guide: - For balanced reasoning, speed, cost, and web search (200k context): `o4-mini` (Pro) - For most powerful reasoning, complex analysis, and web search (200k context): `o3` (Pro) - For flagship performance on complex tasks with very large context (1M tokens): `gpt-4.1` (Pro) - For balanced performance with very large context (1M tokens) at lower cost: `gpt-4.1-mini` (Free) - For most cost-effective option with very large context (1M tokens): `gpt-4.1-nano` (Free) - For comprehensive, high-intelligence tasks with multimodal needs and web search (128k context): `gpt-4o` (Pro) - For high-quality image generation from text or image inputs: `gpt-image-1` (Pro) Practices for Giselle: - Recommended primary model for Pro users: `gpt-4.1` for versatility, capability, intelligence, and web search. - For highest reasoning or very large contexts (up to 200k tokens): `o3` or `o4-mini` (both support web search). - `o3`: Top choice for depth and stability. - `o4-mini`: Strong reasoning with better cost-efficiency. - GPT-4.1 series (`gpt-4.1`, `gpt-4.1-mini`, `gpt-4.1-nano`): Features a ~1 million token context window, improved coding, and instruction following. - Free plan users or cost/speed prioritization for moderate tasks: `gpt-4.1-mini` (1M context) and `gpt-4.1-nano` (1M context). - Image generation: `gpt-image-1` for high-quality results, supporting text and image inputs with quality tiers. - Workflow example: Use `gpt-4o`, `o3`, or `o4-mini` for research with web search, then `o3` for deep analysis, or `gpt-4.1` for extremely large documents. - Official OpenAI Documentation: https://platform.openai.com/docs/models ``` -------------------------------- ### Using the 'all' Toolset Source: https://github.com/github/github-mcp-server Explains how to enable all available toolsets for the github-mcp-server by using the special 'all' toolset identifier, either via command line or environment variable. ```bash ./github-mcp-server --toolsets all ``` ```bash GITHUB_TOOLSETS="all" ./github-mcp-server ``` -------------------------------- ### Prompts Navigation Source: https://docs.giselles.ai/models/providers/anthropic Structures documentation for prompts, including an overview and guides on prompt engineering techniques and examples. ```APIDOC Page Path: /prompts/overview Page Path: /prompts/examples Page Path: /prompts/tips ``` -------------------------------- ### Build GitHub MCP Server from Source Source: https://github.com/github/github-mcp-server Build the GitHub MCP Server binary using the Go toolchain. This process involves navigating to the server's directory and executing the 'go build' command. The resulting executable can then be configured to run with specific arguments and environment variables. ```bash cd cmd/github-mcp-server go build # To specify output location: # go build -o /path/to/your/executable ``` -------------------------------- ### Navbar Links and Primary Action Source: https://docs.giselles.ai/models/providers/anthropic Defines the links in the navigation bar, including support, login, sign up, and a primary 'Get Started' button. ```APIDOC Navbar Links: - Label: Support Href: mailto:support@giselles.ai - Label: Log in Href: https://studio.giselles.ai/login - Label: Sign up Href: https://studio.giselles.ai/signup Navbar Primary Action: Type: button Label: Get Started Href: https://studio.giselles.ai/ ``` -------------------------------- ### Specify Toolsets via Command Line Source: https://github.com/github/github-mcp-server Demonstrates how to specify available toolsets for the github-mcp-server using a command-line argument. ```bash github-mcp-server --toolsets repos,issues,pull_requests,actions,code_security ``` -------------------------------- ### Configure GitHub MCP Server Execution Source: https://github.com/github/github-mcp-server This JSON configuration demonstrates how to set up the GitHub MCP Server to run as a tool. It specifies the command path to the built executable, the arguments to pass (e.g., 'stdio'), and environment variables required for authentication, such as GITHUB_PERSONAL_ACCESS_TOKEN. ```json { "mcp": { "servers": { "github": { "command": "/path/to/github-mcp-server", "args": ["stdio"], "env": { "GITHUB_PERSONAL_ACCESS_TOKEN": "" } } } } } ``` -------------------------------- ### Effective Role Definition Example Source: https://context7_llms Demonstrates how to define a specific role for the AI within a prompt. This guides the AI to produce contextually accurate and relevant outputs for the given task. ```plaintext ✅ Example: "Role: UX Designer Task: Evaluate the provided web interface design for usability issues. Output Format: - List identified issues clearly - Provide specific recommendations for improvements" ``` -------------------------------- ### Get live data with citations example Source: https://docs.perplexity.ai/ Illustrates how to access real-time web information with automatic source attribution and transparency. This snippet focuses on the conceptual implementation of fetching and displaying data with citation. ```html

Get live data with citations

Access real-time web information with automatic source attribution and transparency

``` -------------------------------- ### Perplexity AI: Get Latest News Source: https://docs.perplexity.ai/ Demonstrates how to fetch the latest AI developments using the Perplexity AI chat completions API. This example queries for major AI developments and announcements. ```bash curl https://api.perplexity.ai/chat/completions \ -H "Authorization: Bearer YOUR_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "sonar-pro", "messages": [ { "role": "user", "content": "What are the major AI developments and announcements from today across the tech industry?" } ] }' | jq ``` ```javascript const response = await fetch('https://api.perplexity.ai/chat/completions', { method: 'POST', headers: { 'Authorization': 'Bearer YOUR_API_KEY', 'Content-Type': 'application/json' }, body: JSON.stringify({ model: 'sonar-pro', messages: [ { role: 'user', content: "What are the major AI developments and announcements from today across the tech industry?" } ] }) }); const data = await response.json(); console.log(data); ``` ```python import requests response = requests.post( 'https://api.perplexity.ai/chat/completions', headers={ 'Authorization': 'Bearer YOUR_API_KEY', 'Content-Type': 'application/json' }, json={ 'model': 'sonar-pro', 'messages': [ { 'role': 'user', 'content': "What are the major AI developments and announcements from today across the tech industry?" } ] } ) print(response.json()) ``` -------------------------------- ### GitHub Vector Store Node Setup and Configuration Source: https://context7_llms Instructions for adding and configuring the GitHub Vector Store Node within the Giselle platform. This involves selecting a repository where the Giselle GitHub App is installed to serve as a knowledge source. ```APIDOC GitHub Vector Store Node: __init__() - Adds the node to the Giselle workspace. configure_repository(repository_name: str) - Sets the GitHub repository to be used as a knowledge source. - Requires the Giselle GitHub App to be installed on the repository. - Triggers background vectorization of the repository content. - Status updates from 'REQUIRES SETUP' to repository name upon successful configuration. manage_vector_stores_link() - Provides a link to the team's settings page for managing indexed repositories. - Accessible via Settings > Team > Vector Stores or directly from the node's configuration panel. - URL: https://studio.giselles.ai/settings/team/vector-stores Rate Limits: - Utilizes GitHub App Installation for API access. - Primary rate limit: ~5,000 requests per hour. - Potential for errors with large projects or frequent updates due to rate limiting. - Recommendation: Wait for rate limit window to reset if errors occur. ``` -------------------------------- ### README Content - Giselle Introduction Source: https://github.com/giselles-ai/giselle The introductory section of the Giselle README, presenting the project logo, a Product Hunt badge, and a brief overview of Giselle's purpose. ```html
\ -e GITHUB_TOOLSETS="repos,issues,pull_requests,actions,code_security,experiments" \ ghcr.io/github/github-mcp-server ``` -------------------------------- ### Install Remote GitHub MCP Server in VS Code Insiders Source: https://github.com/github/github-mcp-server Provides a direct link to install the remote GitHub MCP Server in VS Code Insiders. This installation leverages the same remote server hosted by GitHub but is tailored for the Insiders build of VS Code. It includes the essential configuration for the MCP server. ```APIDOC VSCode Insiders MCP Server Installation: URL: https://insiders.vscode.dev/redirect/mcp/install Parameters: name: github config: JSON string specifying server type and URL. - type: "http" - url: "https://api.githubcopilot.com/mcp/" quality: "insiders" Example Configuration: config=%7B%22type%22%3A%20%22http%22%2C%22url%22%3A%20%22https%3A%2F%2Fapi.githubcopilot.com%2Fmcp%2F%22%7D ``` ```JSON { "type": "http", "url": "https://api.githubcopilot.com/mcp/" } ``` -------------------------------- ### Enable Dynamic Toolsets Source: https://github.com/github/github-mcp-server Demonstrates how to enable dynamic toolsets for the GitHub MCP Server. This allows the server to discover and utilize tools dynamically. It shows both command-line execution and Docker environment variable configuration. ```shell ./github-mcp-server --dynamic-toolsets ``` ```docker docker run -i --rm \ -e GITHUB_PERSONAL_ACCESS_TOKEN= \ -e GITHUB_DYNAMIC_TOOLSETS=1 \ ghcr.io/github/github-mcp-server ``` -------------------------------- ### Example SQL Query for User Count Source: https://context7_llms An example of a SQL query that might be generated by an AI agent to count the total number of users in a database table. ```SQL SELECT COUNT(*) FROM users; ``` -------------------------------- ### GitHub Navigation and Product Links Source: https://github.com/github/github-mcp-server Provides links to various GitHub products and features, including Copilot, Models, Advanced Security, Actions, Codespaces, Issues, and more, presented in a navigation menu format. ```text [Skip to content](#start-of-content) Navigation Menu --------------- Toggle navigation [](/) [Sign in](/login?return_to=https%3A%2F%2Fgithub.com%2Fgithub%2Fgithub-mcp-server) Appearance settings * Product * [ GitHub Copilot Write better code with AI ](https://github.com/features/copilot) * [ GitHub Models New Manage and compare prompts ](https://github.com/features/models) * [ GitHub Advanced Security Find and fix vulnerabilities ](https://github.com/security/advanced-security) * [ Actions Automate any workflow ](https://github.com/features/actions) * [ Codespaces Instant dev environments ](https://github.com/features/codespaces) * [ Issues Plan and track work ](https://github.com/features/issues) * [ Code Review Manage code changes ](https://github.com/features/code-review) * [ Discussions Collaborate outside of code ](https://github.com/features/discussions) * [ Code Search Find more, search less ](https://github.com/features/code-search) Explore * [Why GitHub](https://github.com/why-github) * [All features](https://github.com/features) ``` -------------------------------- ### JSON Tool Descriptions Example Source: https://github.com/github/github-mcp-server Demonstrates the JSON format for mapping tool identifiers to their descriptive strings. This structure is used for defining and potentially overriding tool functionalities. ```json { "TOOL_ADD_ISSUE_COMMENT_DESCRIPTION": "an alternative description", "TOOL_CREATE_BRANCH_DESCRIPTION": "Create a new branch in a GitHub repository" } ``` -------------------------------- ### Giselle Generator Node: Technical Query Example (GPT-4o) Source: https://context7_llms An example prompt for Giselle's Generator Nodes, specifically for a technical query using the GPT-4o model, demonstrating effective role definition and output expectations. ```plaintext Role: Experienced programmer Task: Explain the most efficient way to extract unique elements from a Python list. Output Format: - Brief explanation - Python code example demonstrating the method clearly ``` -------------------------------- ### Enable All Toolsets with github-mcp-server Source: https://github.com/github/github-mcp-server This section explains how to enable all available toolsets for the github-mcp-server. You can achieve this by passing the '--toolsets all' argument directly to the server or by setting the GITHUB_TOOLSETS environment variable to 'all'. This ensures all functionalities are accessible. ```shell ./github-mcp-server --toolsets all ``` ```shell GITHUB_TOOLSETS="all" ./github-mcp-server ``` -------------------------------- ### Product Page Launching Today Fragment Source: https://www.producthunt.com/products/giselle A GraphQL fragment detailing information for posts related to products launching today. It includes launch numbers, week status, and references other fragments for vote buttons and page headers. ```graphql fragment ProductPageLaunchingTodayLaunchFragment on Post{ id launchNumber launchedThisWeek launchingToday ...PostVoteButtonFragment ...PostPageHeaderVisitButton ...ProductPageLaunchingTodayFragment } ```