[{"_1":2,"_389":-5,"_390":-5},"loaderData",{"_3":4,"_29":30,"_31":32,"_316":317},"root",{"_5":6,"_14":15},"toasts",[7],{"_8":9,"_10":11,"_12":13},"uid","01972657-a2c9-705b-ab12-1fefd17da646","title","This MCP server does not support inspection","type","error","visitor",{"_16":17,"_18":19},"viewportWidth",1280,"visitorSession",{"_20":21,"_23":24,"_25":-5,"_26":27,"_28":-5},"attributes",[22],"BOT","id",7058884,"membership","referenceId","01971406-43e3-74e8-b880-2f000258007a","userAccount","routes/_public/_layout",{},"routes/_public/mcp/servers/~namespace/~slug/_pages/_layout",{"_33":34,"_87":-5,"_88":47,"_36":89,"_234":235,"_314":315},"availableLocales",[35,47,53,60,67,74,80],{"_36":37,"_40":44,"_45":46},"mcpServer",{"_38":39,"_40":41,"_42":43},"description","Ein erweitertes Integrationspaket, das OpenRouter- und Perplexity-APIs nutzt, um erweiterte Websuchfunktionen bereitzustellen und sowohl einfache als auch komplexe Abfrageverarbeitung mit Unterstützung für Dateianhänge bietet.","name","Perplexity Advanced MCP","readme","[![MseeP.ai-Sicherheitsbewertungsabzeichen](https://mseep.net/pr/code-yeongyu-perplexity-advanced-mcp-badge.png)](https://mseep.ai/app/code-yeongyu-perplexity-advanced-mcp)\n\n# Perplexity Advanced MCP\n\n[![GitHub](https://img.shields.io/badge/GitHub-100000?style=for-the-badge\\&logo=github\\&logoColor=white)](https://github.com/code-yeongyu/perplexity-advanced-mcp) [![PyPI](https://img.shields.io/badge/pypi-3775A9?style=for-the-badge\\&logo=pypi\\&logoColor=white)](https://pypi.org/project/perplexity-advanced-mcp) [![Schmiedeabzeichen](https://smithery.ai/badge/@code-yeongyu/perplexity-advanced-mcp)](https://smithery.ai/server/@code-yeongyu/perplexity-advanced-mcp)\n\n[한국어](README-ko.md)\n\n***\n\n## Überblick\n\nPerplexity Advanced MCP ist ein erweitertes Integrationspaket, das die [OpenRouter-](https://openrouter.ai/) und [Perplexity](https://docs.perplexity.ai/home) -APIs nutzt, um erweiterte Abfrageverarbeitungsfunktionen bereitzustellen. Mit einer intuitiven Befehlszeilenschnittstelle und einem robusten API-Client ermöglicht dieses Paket die nahtlose Interaktion mit KI-Modellen für einfache und komplexe Abfragen.\n\n## Vergleich mit [Perplexity-MCP](https://github.com/jsonallen/perplexity-mcp)\n\nWährend [perplexity-mcp](https://github.com/jsonallen/perplexity-mcp) grundlegende Websuchfunktionen mithilfe der API von [Perplexity](https://docs.perplexity.ai/home) AI bereitstellt, bietet Perplexity Advanced MCP mehrere zusätzliche Funktionen:\n\n* **Multi-Vendor-Support:** Unterstützt sowohl [Perplexity-](https://docs.perplexity.ai/home) als auch [OpenRouter](https://openrouter.ai/) -APIs und bietet Ihnen Flexibilität bei der Wahl Ihres Anbieters\n* **Abfragetypoptimierung:** Unterscheidet zwischen einfachen und komplexen Abfragen und optimiert so Kosten und Leistung\n* **Unterstützung von Dateianhängen:** Ermöglicht das Einbeziehen von Dateiinhalten als Kontext in Ihre Abfragen und ermöglicht so präzisere und kontextbezogenere Antworten\n* **Verbesserte Wiederholungslogik:** Implementiert robuste Wiederholungsmechanismen für verbesserte Zuverlässigkeit\n\nInsgesamt ist dies das am besten geeignete MCP für die Handhabung von Codebasen bei Integration mit Editoren wie [Cline](https://cline.bot/) oder [Cursor](https://www.cursor.com/) .\n\n## Merkmale\n\n* **Einheitlicher API-Client:** Unterstützt sowohl [OpenRouter-](https://openrouter.ai/) als auch [Perplexity](https://docs.perplexity.ai/home) -APIs mit konfigurierbaren Modellen zur Verarbeitung einfacher und komplexer Abfragen.\n* **Befehlszeilenschnittstelle (CLI):** Verwalten Sie die API-Schlüsselkonfiguration und führen Sie den MCP-Server mit [Typer](https://typer.tiangolo.com/) aus.\n* **Erweiterte Abfrageverarbeitung:** Enthält die Verarbeitung von Dateianhängen, sodass Sie Kontextdaten in Ihre Abfragen einbeziehen können.\n* **Robuster Wiederholungsmechanismus:** Nutzt Tenacity für die Wiederholungslogik, um eine konsistente und zuverlässige API-Kommunikation sicherzustellen.\n* **Anpassbare Protokollierung:** Flexible Protokollierungskonfiguration für detailliertes Debuggen und Laufzeitüberwachung.\n\n## Optimale KI-Konfiguration\n\nFür die beste Erfahrung mit KI-Assistenten (z. B. [Cursor](https://www.cursor.com/) , [Claude für Desktop](https://claude.ai/download) ) empfehle ich, Ihren Projektanweisungen oder KI-Regeln die folgende Konfiguration hinzuzufügen:\n\n```xml\n\n \n Perplexity is an LLM that can search the internet, gather information, and answer users' queries.\n\n For example, let's suppose we want to find out the latest version of Python.\n 1. You would search on Google.\n 2. Then read the top two or three results directly to verify.\n\n Perplexity does that work for you.\n\n To answer a user's query, Perplexity searches, opens the top search results, finds information on those websites, and then provides the answer.\n\n Perplexity can be used with two types of queries: simple and complex. Choosing the right query type to fulfill the user's request is most important.\n \n \n \n It's cheap and fast. However, it's not suitable for complex queries. On average, it's more than 10 times cheaper and 3 times faster than complex queries.\n Use it for simple questions such as \"What is the latest version of Python?\"\n \n \n $1/M input tokens\n $1/M output tokens\n \n \n\n \n \n It's slower and more expensive. Compared to simple queries, it's on average more than 10 times more expensive and 3 times slower.\n Use it for more complex requests like \"Analyze the attached code to examine the current status of a specific library and create a migration plan.\"\n \n \n $1/M input tokens\n $5/M output tokens\n \n \n\n \n When reviewing the user's request, if you find anything unexpected, uncertain, or questionable, **and you think you can get answer from the internet**, do not hesitate to use the \"ask_perplexity\" tool to consult Perplexity. However, if the internet is not required to satisfy users' request, it's meaningless to ask to perplexity.\n Since Perplexity is also an LLM, prompt engineering techniques are paramount.\n Remember the basics of prompt engineering, such as providing clear instructions, sufficient context, and examples\n Include as much context and relevant files as possible to smoothly fulfill the user's request. When adding files as attachments, make sure they are absolute paths.\n \n\n```\n\nDiese Konfiguration hilft KI-Assistenten dabei, besser zu verstehen, wann und wie die Perplexity-Suchfunktion verwendet werden soll, und optimiert so sowohl die Kosten als auch die Leistung.\n\n## Verwendung\n\n### Installation über Smithery\n\nSo installieren Sie Perplexity Advanced MCP für Claude Desktop automatisch über [Smithery](https://smithery.ai/server/@code-yeongyu/perplexity-advanced-mcp) :\n\n```bash\nnpx -y @smithery/cli install @code-yeongyu/perplexity-advanced-mcp --client claude\n```\n\n### Schnellstart mit [uvx](https://docs.astral.sh/uv/guides/tools/)\n\nDer einfachste Weg, den MCP-Server auszuführen, ist die Verwendung von [uvx](https://docs.astral.sh/uv/guides/tools/) :\n\n```sh\nuvx perplexity-advanced-mcp -o # or -p \n```\n\nSie können die API-Schlüssel auch mithilfe von Umgebungsvariablen konfigurieren:\n\n```sh\nexport OPENROUTER_API_KEY=\"your_key_here\"\n# or\nexport PERPLEXITY_API_KEY=\"your_key_here\"\n\nuvx perplexity-advanced-mcp\n```\n\nNotiz:\n\n* Die gleichzeitige Angabe von OpenRouter- und Perplexity-API-Schlüsseln führt zu einem Fehler\n* Wenn sowohl CLI-Argumente als auch Umgebungsvariablen angegeben werden, haben CLI-Argumente Vorrang\n\nDie CLI ist mit [Typer](https://typer.tiangolo.com/) erstellt und gewährleistet eine benutzerfreundliche Befehlszeilenerfahrung.\n\n### MCP-Suchtool\n\nDas Paket enthält ein MCP-Suchtool, das über die Funktion `ask_perplexity` integriert ist. Es unterstützt sowohl einfache als auch komplexe Abfragen und verarbeitet Dateianhänge, um zusätzlichen Kontext bereitzustellen.\n\n* **Einfache Abfragen:** Bietet schnelle, effiziente Antworten.\n* **Komplexe Abfragen:** Ermöglicht detailliertes Denken und unterstützt Dateianhänge im XML-Format.\n\n## Konfiguration\n\n* **API-Schlüssel:** Konfigurieren Sie entweder `OPENROUTER_API_KEY` oder `PERPLEXITY_API_KEY` über Befehlszeilenoptionen oder Umgebungsvariablen.\n* **Modellauswahl:** Die Konfiguration (in `src/perplexity_advanced_mcp/config.py` ) ordnet Abfragetypen bestimmten Modellen zu:\n * **[OpenRouter](https://openrouter.ai/) :**\n\n * Einfache Abfragen: `perplexity/sonar`\n * Komplexe Abfragen: `perplexity/sonar-reasoning`\n * **[Ratlosigkeit](https://docs.perplexity.ai/home) :**\n\n * Einfache Abfragen: `sonar-pro`\n * Komplexe Abfragen: `sonar-reasoning-pro`\n\n## Entwicklungshintergrund und -philosophie\n\nDieses Projekt entstand aus meiner persönlichen Neugier und Experimentierfreude. Dem aktuellen [„Vibe-Coding“](https://x.com/karpathy/status/1886192184808149383) -Trend folgend, wurden über 95 % des Codes mit [Cline](https://cline.bot/) + [Cursor](https://www.cursor.com/) IDE geschrieben. Man sagt ja: „Reden ist billig, zeig mir den Code!“ – und mit der Spracherkennungs-Magie von [Wispr Flow](https://wisprflow.ai/) musste ich einfach nur reden, und der Code erschien! Der Großteil der Entwicklung erfolgte durch meine Worte wie „Schreib mir den Code für xyz, behebe den Fehler hier xyz“ und das Drücken der Eingabetaste. Bemerkenswerterweise dauerte die Erstellung dieses voll funktionsfähigen Projekts weniger als ein paar Stunden.\n\nVom Projekt-Scaffolding bis zur Dateistruktur wurde alles über LLM geschrieben und überprüft. Sogar der GitHub Actions-Workflow für die PyPI-Veröffentlichung und der Release-Freigabeprozess wurden über Cursor abgewickelt. Als menschlicher Entwickler bestand meine Aufgabe darin:\n\n* Starten und Stoppen des MCP-Servers, um der KI die Durchführung ordnungsgemäßer Tests zu erleichtern\n* Kopieren und Bereitstellen von Fehlerprotokollen bei aufgetretenen Problemen\n* Suchen und Bereitstellen [von Python MCP SDK](https://github.com/modelcontextprotocol/python-sdk) -Dokumentation und Beispielen aus dem Internet\n* Anfordern von Änderungen für Code, der nicht korrekt erscheint\n\nIn der heutigen Welt, in der vieles automatisiert und ersetzt werden kann, hoffe ich, dass dieses MCP Entwicklern wie Ihnen, die es nutzen, dabei helfen kann, Mehrwerte zu entdecken, die über das bloße Schreiben von Code hinausgehen. Möge dieses Tool Sie dabei unterstützen, ein Entwickler der neuen Ära zu werden, der Entscheidungen und Überlegungen auf höherer Ebene treffen kann.\n\n## Entwicklung\n\nSo tragen Sie zu diesem Paket bei oder ändern es:\n\n### 1. **Klonen Sie das Repository:**\n\n```sh\ngh repo clone code-yeongyu/perplexity-advanced-mcp\n```\n\n### 2. **Abhängigkeiten installieren:**\n\n```sh\nuv sync\n```\n\n### 3. **Beitragen:**\n\nBeiträge sind willkommen! Bitte beachten Sie den bestehenden Codestil und die Commit-Richtlinien.\n\n## Lizenz\n\nDieses Projekt ist unter der MIT-Lizenz lizenziert.","German","tag","de-DE",{"_36":48,"_40":51,"_45":52},{"_38":49,"_40":41,"_42":50},"An advanced integration package that leverages OpenRouter and Perplexity APIs to provide enhanced web search capabilities, offering both simple and complex query processing with file attachment support.","[![MseeP.ai Security Assessment Badge](https://mseep.net/pr/code-yeongyu-perplexity-advanced-mcp-badge.png)](https://mseep.ai/app/code-yeongyu-perplexity-advanced-mcp)\n\n

\n\n# Perplexity Advanced MCP\n\n[![GitHub](https://img.shields.io/badge/GitHub-100000?style=for-the-badge&logo=github&logoColor=white)](https://github.com/code-yeongyu/perplexity-advanced-mcp)\n[![PyPI](https://img.shields.io/badge/pypi-3775A9?style=for-the-badge&logo=pypi&logoColor=white)](https://pypi.org/project/perplexity-advanced-mcp)\n[![smithery badge](https://smithery.ai/badge/@code-yeongyu/perplexity-advanced-mcp)](https://smithery.ai/server/@code-yeongyu/perplexity-advanced-mcp)\n\n[한국어](README-ko.md)\n\n

\n\n---\n\n## Overview\n\nPerplexity Advanced MCP is an advanced integration package that leverages the [OpenRouter](https://openrouter.ai/) and [Perplexity](https://docs.perplexity.ai/home) APIs to provide enhanced query processing capabilities. With an intuitive command-line interface and a robust API client, this package facilitates seamless interactions with AI models for both simple and complex queries.\n\n## Comparison with [perplexity-mcp](https://github.com/jsonallen/perplexity-mcp)\n\nWhile [perplexity-mcp](https://github.com/jsonallen/perplexity-mcp) provides basic web search functionality using [Perplexity](https://docs.perplexity.ai/home) AI's API, Perplexity Advanced MCP offers several additional features:\n\n- **Multi-vendor Support:** Supports both [Perplexity](https://docs.perplexity.ai/home) and [OpenRouter](https://openrouter.ai/) APIs, giving you flexibility in choosing your provider\n- **Query Type Optimization:** Distinguishes between simple and complex queries, optimizing for cost and performance\n- **File Attachment Support:** Allows including file contents as context in your queries, enabling more precise and contextual responses\n- **Enhanced Retry Logic:** Implements robust retry mechanisms for improved reliability\n\nOverall, this is the most suitable MCP for handling codebases when integrated with editors like [Cline](https://cline.bot/) or [Cursor](https://www.cursor.com/).\n\n\n## Features\n\n- **Unified API Client:** Supports both [OpenRouter](https://openrouter.ai/) and [Perplexity](https://docs.perplexity.ai/home) APIs with configurable models for handling simple and complex queries.\n- **Command-Line Interface (CLI):** Manage API key configuration and run the MCP server using [Typer](https://typer.tiangolo.com/).\n- **Advanced Query Processing:** Incorporates file attachment processing, allowing you to include contextual data in your queries.\n- **Robust Retry Mechanism:** Utilizes Tenacity for retry logic to ensure consistent and reliable API communications.\n- **Customizable Logging:** Flexible logging configuration for detailed debugging and runtime monitoring.\n\n## Optimal AI Configuration\n\nFor the best experience with AI assistants (e.g., [Cursor](https://www.cursor.com/), [Claude for Desktop](https://claude.ai/download)), I recommend adding the following configuration to your project instructions or AI rules:\n\n```xml\n\n \n Perplexity is an LLM that can search the internet, gather information, and answer users' queries.\n\n For example, let's suppose we want to find out the latest version of Python.\n 1. You would search on Google.\n 2. Then read the top two or three results directly to verify.\n\n Perplexity does that work for you.\n\n To answer a user's query, Perplexity searches, opens the top search results, finds information on those websites, and then provides the answer.\n\n Perplexity can be used with two types of queries: simple and complex. Choosing the right query type to fulfill the user's request is most important.\n \n \n \n It's cheap and fast. However, it's not suitable for complex queries. On average, it's more than 10 times cheaper and 3 times faster than complex queries.\n Use it for simple questions such as \"What is the latest version of Python?\"\n \n \n $1/M input tokens\n $1/M output tokens\n \n \n\n \n \n It's slower and more expensive. Compared to simple queries, it's on average more than 10 times more expensive and 3 times slower.\n Use it for more complex requests like \"Analyze the attached code to examine the current status of a specific library and create a migration plan.\"\n \n \n $1/M input tokens\n $5/M output tokens\n \n \n\n \n When reviewing the user's request, if you find anything unexpected, uncertain, or questionable, **and you think you can get answer from the internet**, do not hesitate to use the \"ask_perplexity\" tool to consult Perplexity. However, if the internet is not required to satisfy users' request, it's meaningless to ask to perplexity.\n Since Perplexity is also an LLM, prompt engineering techniques are paramount.\n Remember the basics of prompt engineering, such as providing clear instructions, sufficient context, and examples\n Include as much context and relevant files as possible to smoothly fulfill the user's request. When adding files as attachments, make sure they are absolute paths.\n \n\n```\n\nThis configuration helps AI assistants better understand when and how to use the Perplexity search functionality, optimizing for both cost and performance.\n\n## Usage\n\n### Installing via Smithery\n\nTo install Perplexity Advanced MCP for Claude Desktop automatically via [Smithery](https://smithery.ai/server/@code-yeongyu/perplexity-advanced-mcp):\n\n```bash\nnpx -y @smithery/cli install @code-yeongyu/perplexity-advanced-mcp --client claude\n```\n\n### Quick Start with [uvx](https://docs.astral.sh/uv/guides/tools/)\n\nThe easiest way to run the MCP server is using [uvx](https://docs.astral.sh/uv/guides/tools/):\n\n```sh\nuvx perplexity-advanced-mcp -o # or -p \n```\n\nYou can also configure the API keys using environment variables:\n\n```sh\nexport OPENROUTER_API_KEY=\"your_key_here\"\n# or\nexport PERPLEXITY_API_KEY=\"your_key_here\"\n\nuvx perplexity-advanced-mcp\n```\n\nNote:\n- Providing both OpenRouter and Perplexity API keys simultaneously will result in an error\n- When both CLI arguments and environment variables are provided, CLI arguments take precedence\n\nThe CLI is built with [Typer](https://typer.tiangolo.com/), ensuring a user-friendly command-line experience.\n\n### MCP Search Tool\n\nThe package includes an MCP search tool integrated via the `ask_perplexity` function. It supports both simple and complex queries and processes file attachments to provide additional context.\n\n- **Simple Queries:** Provides fast, efficient responses.\n- **Complex Queries:** Engages in detailed reasoning and supports file attachments formatted as XML.\n\n## Configuration\n\n- **API Keys:** Configure either the `OPENROUTER_API_KEY` or `PERPLEXITY_API_KEY` through command-line options or environment variables.\n- **Model Selection:** The configuration (in `src/perplexity_advanced_mcp/config.py`) maps query types to specific models:\n - **[OpenRouter](https://openrouter.ai/):**\n - Simple Queries: `perplexity/sonar`\n - Complex Queries: `perplexity/sonar-reasoning`\n - **[Perplexity](https://docs.perplexity.ai/home):**\n - Simple Queries: `sonar-pro`\n - Complex Queries: `sonar-reasoning-pro`\n\n## Development Background & Philosophy\n\nThis project emerged from my personal curiosity and experimentation. Following the recent [\"vibe coding\"](https://x.com/karpathy/status/1886192184808149383) trend, over 95% of the code was written through [Cline](https://cline.bot/) + [Cursor](https://www.cursor.com/) IDE. They say \"talk is cheap, show me the code\" - well, with [Wispr Flow](https://wisprflow.ai/)'s speech-to-text magic, I literally just talked and the code showed up! Most of the development was done by me saying things like \"Write me the code for x y z, fix the bug here x y z.\" and pressing enter. Remarkably, creating this fully functional project took less than a few hours.\n\nFrom project scaffolding to file structure, everything was written and reviewed through LLM. Even the GitHub Actions workflow for PyPI publishing and the release approval process were handled through Cursor. As a human developer, my role was to:\n\n- Starting and stopping the MCP server to help AI conduct proper testing\n- Copying and providing error logs when issues occurred\n- Finding and providing [Python MCP SDK](https://github.com/modelcontextprotocol/python-sdk) documentation and examples from the internet\n- Requesting modifications for code that didn't seem correct\n\nIn today's world where many things can be automated and replaced, I hope this MCP can help developers like you who use it to discover value beyond just writing code. May this tool assist you in becoming a new era developer who can make higher-level decisions and considerations.\n\n## Development\n\nTo contribute or modify this package:\n\n### 1. **Clone the Repository:**\n\n```sh\ngh repo clone code-yeongyu/perplexity-advanced-mcp\n```\n\n### 2. **Install Dependencies:**\n\n```sh\nuv sync\n```\n\n### 3. **Contribute:**\n\nContributions are welcome! Please follow the existing code style and commit guidelines.\n\n## License\n\nThis project is licensed under the MIT License.\n","English","en-US",{"_36":54,"_40":58,"_45":59},{"_38":55,"_40":56,"_42":57},"Un paquete de integración avanzado que aprovecha las API de OpenRouter y Perplexity para proporcionar capacidades de búsqueda web mejoradas, ofreciendo procesamiento de consultas simples y complejas con soporte para adjuntar archivos.","Perplejidad avanzada MCP","# Perplejidad avanzada MCP\n\n[![GitHub](https://img.shields.io/badge/GitHub-100000?style=for-the-badge\\&logo=github\\&logoColor=white)](https://github.com/code-yeongyu/perplexity-advanced-mcp) [![PyPI](https://img.shields.io/badge/pypi-3775A9?style=for-the-badge\\&logo=pypi\\&logoColor=white)](https://pypi.org/project/perplexity-advanced-mcp) [![insignia de herrería](https://smithery.ai/badge/@code-yeongyu/perplexity-advanced-mcp)](https://smithery.ai/server/@code-yeongyu/perplexity-advanced-mcp)\n\n[Coreano](README-ko.md)\n\n***\n\n## Descripción general\n\nPerplexity Advanced MCP es un paquete de integración avanzado que aprovecha las API de [OpenRouter](https://openrouter.ai/) y [Perplexity](https://docs.perplexity.ai/home) para ofrecer capacidades mejoradas de procesamiento de consultas. Con una interfaz de línea de comandos intuitiva y un cliente API robusto, este paquete facilita interacciones fluidas con modelos de IA para consultas simples y complejas.\n\n## Comparación con [perplejidad-mcp](https://github.com/jsonallen/perplexity-mcp)\n\nSi bien [perplexity-mcp](https://github.com/jsonallen/perplexity-mcp) proporciona una funcionalidad básica de búsqueda web mediante la API de [Perplexity](https://docs.perplexity.ai/home) AI, Perplexity Advanced MCP ofrece varias funciones adicionales:\n\n* **Compatibilidad con múltiples proveedores:** admite las API de [Perplexity](https://docs.perplexity.ai/home) y [OpenRouter](https://openrouter.ai/) , lo que le brinda flexibilidad para elegir su proveedor.\n* **Optimización del tipo de consulta:** distingue entre consultas simples y complejas, optimizando el costo y el rendimiento.\n* **Compatibilidad con archivos adjuntos:** permite incluir contenidos de archivos como contexto en sus consultas, lo que permite respuestas más precisas y contextuales.\n* **Lógica de reintento mejorada:** implementa mecanismos de reintento robustos para una mayor confiabilidad\n\nEn general, este es el MCP más adecuado para manejar bases de código cuando se integra con editores como [Cline](https://cline.bot/) o [Cursor](https://www.cursor.com/) .\n\n## Características\n\n* **Cliente API unificado:** admite las API de [OpenRouter](https://openrouter.ai/) y [Perplexity](https://docs.perplexity.ai/home) con modelos configurables para manejar consultas simples y complejas.\n* **Interfaz de línea de comandos (CLI):** administre la configuración de la clave API y ejecute el servidor MCP mediante [Typer](https://typer.tiangolo.com/) .\n* **Procesamiento avanzado de consultas:** incorpora procesamiento de archivos adjuntos, lo que le permite incluir datos contextuales en sus consultas.\n* **Mecanismo de reintento robusto:** utiliza Tenacity para la lógica de reintento para garantizar comunicaciones de API consistentes y confiables.\n* **Registro personalizable:** configuración de registro flexible para una depuración detallada y monitoreo del tiempo de ejecución.\n\n## Configuración óptima de IA\n\nPara obtener la mejor experiencia con los asistentes de IA (por ejemplo, [Cursor](https://www.cursor.com/) , [Claude for Desktop](https://claude.ai/download) ), recomiendo agregar la siguiente configuración a las instrucciones de su proyecto o reglas de IA:\n\n```xml\n\n \n Perplexity is an LLM that can search the internet, gather information, and answer users' queries.\n\n For example, let's suppose we want to find out the latest version of Python.\n 1. You would search on Google.\n 2. Then read the top two or three results directly to verify.\n\n Perplexity does that work for you.\n\n To answer a user's query, Perplexity searches, opens the top search results, finds information on those websites, and then provides the answer.\n\n Perplexity can be used with two types of queries: simple and complex. Choosing the right query type to fulfill the user's request is most important.\n \n \n \n It's cheap and fast. However, it's not suitable for complex queries. On average, it's more than 10 times cheaper and 3 times faster than complex queries.\n Use it for simple questions such as \"What is the latest version of Python?\"\n \n \n $1/M input tokens\n $1/M output tokens\n \n \n\n \n \n It's slower and more expensive. Compared to simple queries, it's on average more than 10 times more expensive and 3 times slower.\n Use it for more complex requests like \"Analyze the attached code to examine the current status of a specific library and create a migration plan.\"\n \n \n $1/M input tokens\n $5/M output tokens\n \n \n\n \n When reviewing the user's request, if you find anything unexpected, uncertain, or questionable, **and you think you can get answer from the internet**, do not hesitate to use the \"ask_perplexity\" tool to consult Perplexity. However, if the internet is not required to satisfy users' request, it's meaningless to ask to perplexity.\n Since Perplexity is also an LLM, prompt engineering techniques are paramount.\n Remember the basics of prompt engineering, such as providing clear instructions, sufficient context, and examples\n Include as much context and relevant files as possible to smoothly fulfill the user's request. When adding files as attachments, make sure they are absolute paths.\n \n\n```\n\nEsta configuración ayuda a los asistentes de IA a comprender mejor cuándo y cómo utilizar la funcionalidad de búsqueda de Perplexity, optimizando tanto el costo como el rendimiento.\n\n## Uso\n\n### Instalación mediante herrería\n\nPara instalar Perplexity Advanced MCP para Claude Desktop automáticamente a través de [Smithery](https://smithery.ai/server/@code-yeongyu/perplexity-advanced-mcp) :\n\n```bash\nnpx -y @smithery/cli install @code-yeongyu/perplexity-advanced-mcp --client claude\n```\n\n### Inicio rápido con [uvx](https://docs.astral.sh/uv/guides/tools/)\n\nLa forma más sencilla de ejecutar el servidor MCP es usando [uvx](https://docs.astral.sh/uv/guides/tools/) :\n\n```sh\nuvx perplexity-advanced-mcp -o # or -p \n```\n\nTambién puede configurar las claves API mediante variables de entorno:\n\n```sh\nexport OPENROUTER_API_KEY=\"your_key_here\"\n# or\nexport PERPLEXITY_API_KEY=\"your_key_here\"\n\nuvx perplexity-advanced-mcp\n```\n\nNota:\n\n* Proporcionar claves API de OpenRouter y Perplexity simultáneamente generará un error\n* Cuando se proporcionan argumentos CLI y variables de entorno, los argumentos CLI tienen prioridad\n\nLa CLI está construida con [Typer](https://typer.tiangolo.com/) , lo que garantiza una experiencia de línea de comandos fácil de usar.\n\n### Herramienta de búsqueda de MCP\n\nEl paquete incluye una herramienta de búsqueda MCP integrada mediante la función `ask_perplexity` . Admite consultas simples y complejas y procesa archivos adjuntos para proporcionar contexto adicional.\n\n* **Consultas simples:** proporciona respuestas rápidas y eficientes.\n* **Consultas complejas:** realiza razonamientos detallados y admite archivos adjuntos con formato XML.\n\n## Configuración\n\n* **Claves API:** configure `OPENROUTER_API_KEY` o `PERPLEXITY_API_KEY` a través de opciones de línea de comandos o variables de entorno.\n* **Selección de modelo:** la configuración (en `src/perplexity_advanced_mcp/config.py` ) asigna tipos de consulta a modelos específicos:\n * **[OpenRouter](https://openrouter.ai/) :**\n\n * Consultas simples: `perplexity/sonar`\n * Consultas complejas: `perplexity/sonar-reasoning`\n * **[Perplejidad](https://docs.perplexity.ai/home) :**\n\n * Consultas simples: `sonar-pro`\n * Consultas complejas: `sonar-reasoning-pro`\n\n## Antecedentes y filosofía del desarrollo\n\nEste proyecto surgió de mi curiosidad y experimentación personal. Siguiendo la reciente tendencia [de \"codificación vibrante\"](https://x.com/karpathy/status/1886192184808149383) , más del 95% del código se escribió con [Cline](https://cline.bot/) + [Cursor](https://www.cursor.com/) IDE. Dicen que \"hablar es fácil, enséñame el código\"; bueno, con la magia de voz a texto de [Wispr Flow](https://wisprflow.ai/) , ¡literalmente solo hablé y el código apareció! La mayor parte del desarrollo se realizó diciendo cosas como \"Escríbeme el código para xyz, corrige el error aquí xy z\" y presionando Enter. Sorprendentemente, crear este proyecto completamente funcional me llevó menos de unas horas.\n\nDesde el andamiaje del proyecto hasta la estructura de archivos, todo se escribió y revisó mediante LLM. Incluso el flujo de trabajo de GitHub Actions para la publicación en PyPI y el proceso de aprobación de versiones se gestionaron mediante Cursor. Como desarrolladora, mi función era:\n\n* Iniciar y detener el servidor MCP para ayudar a la IA a realizar pruebas adecuadas\n* Copiar y proporcionar registros de errores cuando ocurrieron problemas\n* Encontrar y proporcionar documentación y ejemplos [del SDK de Python MCP](https://github.com/modelcontextprotocol/python-sdk) en Internet\n* Solicitar modificaciones para el código que no parecía correcto\n\nEn el mundo actual, donde muchas cosas se pueden automatizar y reemplazar, espero que este MCP ayude a desarrolladores como tú a descubrir valor más allá de escribir código. Espero que esta herramienta te ayude a convertirte en un desarrollador de la nueva era capaz de tomar decisiones y reflexionar sobre temas de mayor nivel.\n\n## Desarrollo\n\nPara contribuir o modificar este paquete:\n\n### 1. **Clonar el repositorio:**\n\n```sh\ngh repo clone code-yeongyu/perplexity-advanced-mcp\n```\n\n### 2. **Instalar dependencias:**\n\n```sh\nuv sync\n```\n\n### 3. **Contribuir:**\n\n¡Agradecemos sus contribuciones! Por favor, sigan las pautas de estilo y envío de código existentes.\n\n## Licencia\n\nEste proyecto está licenciado bajo la licencia MIT.","Spanish","es-ES",{"_36":61,"_40":65,"_45":66},{"_38":62,"_40":63,"_42":64},"OpenRouter および Perplexity API を活用して強化された Web 検索機能を提供し、ファイル添付のサポートによる単純なクエリ処理と複雑なクエリ処理の両方を提供する高度な統合パッケージです。","困惑の高度なMCP","# 困惑の高度なMCP\n\n[![GitHub](https://img.shields.io/badge/GitHub-100000?style=for-the-badge\\&logo=github\\&logoColor=white)](https://github.com/code-yeongyu/perplexity-advanced-mcp) [![パイパイ](https://img.shields.io/badge/pypi-3775A9?style=for-the-badge\\&logo=pypi\\&logoColor=white)](https://pypi.org/project/perplexity-advanced-mcp) [![鍛冶屋のバッジ](https://smithery.ai/badge/@code-yeongyu/perplexity-advanced-mcp)](https://smithery.ai/server/@code-yeongyu/perplexity-advanced-mcp)\n\n[韓国語](README-ko.md)\n\n***\n\n## 概要\n\nPerplexity Advanced MCPは、 [OpenRouter](https://openrouter.ai/)および[Perplexity](https://docs.perplexity.ai/home) APIを活用し、強化されたクエリ処理機能を提供する高度な統合パッケージです。直感的なコマンドラインインターフェースと堅牢なAPIクライアントを備えたこのパッケージは、単純なクエリから複雑なクエリまで、AIモデルとのシームレスな連携を実現します。\n\n## [perplexity-mcp](https://github.com/jsonallen/perplexity-mcp)との比較\n\n[perplexity-mcp は](https://github.com/jsonallen/perplexity-mcp)[Perplexity](https://docs.perplexity.ai/home) AI の API を使用して基本的な Web 検索機能を提供しますが、Perplexity Advanced MCP はいくつかの追加機能を提供します。\n\n* **マルチベンダーサポート:** [Perplexity](https://docs.perplexity.ai/home) APIと[OpenRouter](https://openrouter.ai/) APIの両方をサポートし、プロバイダーを柔軟に選択できます。\n* \\*\\*クエリタイプの最適化:\\*\\*単純なクエリと複雑なクエリを区別し、コストとパフォーマンスを最適化します。\n* \\*\\*ファイル添付のサポート:\\*\\*クエリにファイルの内容をコンテキストとして含めることができるため、より正確でコンテキストに基づいた応答が可能になります。\n* \\*\\*強化された再試行ロジック:\\*\\*信頼性を向上させる堅牢な再試行メカニズムを実装します\n\n全体として、これは[Cline](https://cline.bot/)や[Cursor](https://www.cursor.com/)などのエディターと統合されたときにコードベースを処理するのに最も適した MCP です。\n\n## 特徴\n\n* \\*\\*統合 API クライアント:\\*\\*単純なクエリと複雑なクエリを処理するための構成可能なモデルを備えた[OpenRouter](https://openrouter.ai/) API と[Perplexity](https://docs.perplexity.ai/home) API の両方をサポートします。\n* **コマンドラインインターフェイス (CLI):** [Typer](https://typer.tiangolo.com/)を使用して API キー構成を管理し、MCP サーバーを実行します。\n* \\*\\*高度なクエリ処理:\\*\\*ファイル添付処理を組み込んでおり、クエリにコンテキストデータを含めることができます。\n* \\*\\*堅牢な再試行メカニズム:\\*\\*再試行ロジックに Tenacity を利用して、一貫性と信頼性の高い API 通信を確保します。\n* \\*\\*カスタマイズ可能なログ記録:\\*\\*詳細なデバッグと実行時監視のための柔軟なログ記録構成。\n\n## 最適なAI構成\n\nAI アシスタント (例: [Cursor](https://www.cursor.com/) 、 [Claude for Desktop](https://claude.ai/download) ) を最大限に活用するには、プロジェクトの指示または AI ルールに次の構成を追加することをお勧めします。\n\n```xml\n\n \n Perplexity is an LLM that can search the internet, gather information, and answer users' queries.\n\n For example, let's suppose we want to find out the latest version of Python.\n 1. You would search on Google.\n 2. Then read the top two or three results directly to verify.\n\n Perplexity does that work for you.\n\n To answer a user's query, Perplexity searches, opens the top search results, finds information on those websites, and then provides the answer.\n\n Perplexity can be used with two types of queries: simple and complex. Choosing the right query type to fulfill the user's request is most important.\n \n \n \n It's cheap and fast. However, it's not suitable for complex queries. On average, it's more than 10 times cheaper and 3 times faster than complex queries.\n Use it for simple questions such as \"What is the latest version of Python?\"\n \n \n $1/M input tokens\n $1/M output tokens\n \n \n\n \n \n It's slower and more expensive. Compared to simple queries, it's on average more than 10 times more expensive and 3 times slower.\n Use it for more complex requests like \"Analyze the attached code to examine the current status of a specific library and create a migration plan.\"\n \n \n $1/M input tokens\n $5/M output tokens\n \n \n\n \n When reviewing the user's request, if you find anything unexpected, uncertain, or questionable, **and you think you can get answer from the internet**, do not hesitate to use the \"ask_perplexity\" tool to consult Perplexity. However, if the internet is not required to satisfy users' request, it's meaningless to ask to perplexity.\n Since Perplexity is also an LLM, prompt engineering techniques are paramount.\n Remember the basics of prompt engineering, such as providing clear instructions, sufficient context, and examples\n Include as much context and relevant files as possible to smoothly fulfill the user's request. When adding files as attachments, make sure they are absolute paths.\n \n\n```\n\nこの構成により、AI アシスタントは Perplexity 検索機能をいつどのように使用するかをより適切に理解できるようになり、コストとパフォーマンスの両方が最適化されます。\n\n## 使用法\n\n### Smithery経由でインストール\n\n[Smithery](https://smithery.ai/server/@code-yeongyu/perplexity-advanced-mcp)経由で Claude Desktop 用の Perplexity Advanced MCP を自動的にインストールするには:\n\n```bash\nnpx -y @smithery/cli install @code-yeongyu/perplexity-advanced-mcp --client claude\n```\n\n### [uvx](https://docs.astral.sh/uv/guides/tools/)のクイックスタート\n\nMCP サーバーを実行する最も簡単な方法は、 [uvx](https://docs.astral.sh/uv/guides/tools/)を使用することです。\n\n```sh\nuvx perplexity-advanced-mcp -o # or -p \n```\n\n環境変数を使用して API キーを構成することもできます。\n\n```sh\nexport OPENROUTER_API_KEY=\"your_key_here\"\n# or\nexport PERPLEXITY_API_KEY=\"your_key_here\"\n\nuvx perplexity-advanced-mcp\n```\n\n注記：\n\n* OpenRouterとPerplexity APIキーを同時に提供するとエラーが発生します。\n* CLI引数と環境変数の両方が指定されている場合は、CLI引数が優先されます。\n\nCLI は[Typer](https://typer.tiangolo.com/)を使用して構築されており、ユーザーフレンドリーなコマンドラインエクスペリエンスを保証します。\n\n### MCP検索ツール\n\nこのパッケージには、 `ask_perplexity`関数を介して統合されたMCP検索ツールが含まれています。単純なクエリと複雑なクエリの両方をサポートし、添付ファイルを処理して追加のコンテキストを提供します。\n\n* \\*\\*シンプルなクエリ:\\*\\*高速で効率的な応答を提供します。\n* \\*\\*複雑なクエリ:\\*\\*詳細な推論を実行し、XML 形式のファイル添付をサポートします。\n\n## 構成\n\n* \\*\\*API キー:\\*\\*コマンドラインオプションまたは環境変数を使用して、 `OPENROUTER_API_KEY`または`PERPLEXITY_API_KEY`いずれかを構成します。\n* \\*\\*モデルの選択:\\*\\*構成 ( `src/perplexity_advanced_mcp/config.py`内) は、クエリタイプを特定のモデルにマッピングします。\n * **[オープンルーター](https://openrouter.ai/):**\n\n * シンプルなクエリ: `perplexity/sonar`\n * 複雑なクエリ: `perplexity/sonar-reasoning`\n * **[困惑](https://docs.perplexity.ai/home)：**\n\n * シンプルなクエリ: `sonar-pro`\n * 複雑なクエリ: `sonar-reasoning-pro`\n\n## 開発の背景と理念\n\nこのプロジェクトは、私の個人的な好奇心と実験から生まれました。最近の[「バイブコーディング」の](https://x.com/karpathy/status/1886192184808149383)トレンドに倣い、コードの95%以上を[Cline](https://cline.bot/) + [Cursor](https://www.cursor.com/) IDEで作成しました。「話すのは簡単、コードを見せて」と言われますが、 [Wispr Flow](https://wisprflow.ai/)の音声テキスト変換機能を使えば、文字通り話しかけるだけでコードが表示されました！開発の大部分は、「xyzのコードを書いて、ここのバグをxy zに修正して」などと指示してEnterキーを押すだけで完了しました。驚くべきことに、この完全に機能するプロジェクトの作成には数時間もかかりませんでした。\n\nプロジェクトのスキャフォールディングからファイル構造まで、すべてLLMで作成・レビューされました。PyPI公開のためのGitHub Actionsワークフローやリリース承認プロセスもCursorで処理されました。人間の開発者としての私の役割は以下のとおりです。\n\n* AIが適切なテストを実施できるようにMCPサーバーの起動と停止を行う\n* 問題が発生した場合はエラーログをコピーして提供する\n* インターネットから[Python MCP SDKの](https://github.com/modelcontextprotocol/python-sdk)ドキュメントとサンプルを見つけて提供する\n* 正しくないと思われるコードの変更を依頼する\n\n多くのものが自動化され、代替される今日の世界において、このMCPが、コードを書く以上の価値を見出すために活用する皆さんのような開発者の助けとなることを願っています。このツールが、より高度な意思決定と検討ができる新時代の開発者へと成長するための一助となりますように。\n\n## 発達\n\nこのパッケージに貢献または変更するには:\n\n### 1.**リポジトリのクローンを作成します。**\n\n```sh\ngh repo clone code-yeongyu/perplexity-advanced-mcp\n```\n\n### 2.**依存関係をインストールします。**\n\n```sh\nuv sync\n```\n\n### 3.**貢献する:**\n\n貢献を歓迎します！既存のコードスタイルとコミットガイドラインに従ってください。\n\n## ライセンス\n\nこのプロジェクトは MIT ライセンスに基づいてライセンスされています。","Japanese","ja-JP",{"_36":68,"_40":72,"_45":73},{"_38":69,"_40":70,"_42":71},"OpenRouter와 Perplexity API를 활용하여 향상된 웹 검색 기능을 제공하는 고급 통합 패키지로, 파일 첨부 지원을 통해 간단한 쿼리 처리와 복잡한 쿼리 처리를 모두 제공합니다.","Perplexity 고급 MCP","# Perplexity 고급 MCP\n\n[![깃허브](https://img.shields.io/badge/GitHub-100000?style=for-the-badge\\&logo=github\\&logoColor=white)](https://github.com/code-yeongyu/perplexity-advanced-mcp) [![파이파이](https://img.shields.io/badge/pypi-3775A9?style=for-the-badge\\&logo=pypi\\&logoColor=white)](https://pypi.org/project/perplexity-advanced-mcp) [![대장간 배지](https://smithery.ai/badge/@code-yeongyu/perplexity-advanced-mcp)](https://smithery.ai/server/@code-yeongyu/perplexity-advanced-mcp)\n\n[한국어](README-ko.md)\n\n***\n\n## 개요\n\nPerplexity Advanced MCP는 [OpenRouter](https://openrouter.ai/) 및 [Perplexity](https://docs.perplexity.ai/home) API를 활용하여 향상된 쿼리 처리 기능을 제공하는 고급 통합 패키지입니다. 직관적인 명령줄 인터페이스와 강력한 API 클라이언트를 갖춘 이 패키지는 간단한 쿼리부터 복잡한 쿼리까지 AI 모델과의 원활한 상호작용을 지원합니다.\n\n## [perplexity-mcp](https://github.com/jsonallen/perplexity-mcp) 와의 비교\n\n[perplexity-mcp는](https://github.com/jsonallen/perplexity-mcp) [Perplexity](https://docs.perplexity.ai/home) AI의 API를 사용하여 기본적인 웹 검색 기능을 제공하는 반면, Perplexity Advanced MCP는 여러 가지 추가 기능을 제공합니다.\n\n* **다중 공급업체 지원:** [Perplexity](https://docs.perplexity.ai/home) 와 [OpenRouter](https://openrouter.ai/) API를 모두 지원하므로 공급업체를 선택할 때 유연성이 제공됩니다.\n* **쿼리 유형 최적화:** 간단한 쿼리와 복잡한 쿼리를 구분하여 비용과 성능을 최적화합니다.\n* **파일 첨부 지원:** 쿼리에 파일 내용을 컨텍스트로 포함시켜 보다 정확하고 상황에 맞는 응답을 제공할 수 있습니다.\n* **향상된 재시도 논리:** 향상된 안정성을 위해 강력한 재시도 메커니즘을 구현합니다.\n\n전반적으로 이것은 [Cline](https://cline.bot/) 이나 [Cursor와](https://www.cursor.com/) 같은 편집기와 통합할 때 코드베이스를 처리하는 데 가장 적합한 MCP입니다.\n\n## 특징\n\n* **통합 API 클라이언트:** 간단한 쿼리와 복잡한 쿼리를 처리하기 위한 구성 가능한 모델을 통해 [OpenRouter](https://openrouter.ai/) 와 [Perplexity](https://docs.perplexity.ai/home) API를 모두 지원합니다.\n* **명령줄 인터페이스(CLI):** [Typer를](https://typer.tiangolo.com/) 사용하여 API 키 구성을 관리하고 MCP 서버를 실행합니다.\n* **고급 쿼리 처리:** 파일 첨부 처리를 통합하여 쿼리에 문맥적 데이터를 포함할 수 있습니다.\n* **강력한 재시도 메커니즘:** 일관되고 안정적인 API 통신을 보장하기 위해 재시도 논리에 Tenacity를 활용합니다.\n* **사용자 정의 로깅:** 세부적인 디버깅 및 런타임 모니터링을 위한 유연한 로깅 구성.\n\n## 최적의 AI 구성\n\nAI 어시스턴트(예: [Cursor](https://www.cursor.com/) , [Claude for Desktop](https://claude.ai/download) )를 가장 잘 활용하려면 프로젝트 지침이나 AI 규칙에 다음 구성을 추가하는 것이 좋습니다.\n\n지엑스피1\n\n이러한 구성은 AI 도우미가 Perplexity 검색 기능을 언제 어떻게 사용해야 하는지 더 잘 이해하고 비용과 성능을 모두 최적화하는 데 도움이 됩니다.\n\n## 용법\n\n### Smithery를 통해 설치\n\n[Smithery를](https://smithery.ai/server/@code-yeongyu/perplexity-advanced-mcp) 통해 Claude Desktop용 Perplexity Advanced MCP를 자동으로 설치하려면:\n\n```bash\nnpx -y @smithery/cli install @code-yeongyu/perplexity-advanced-mcp --client claude\n```\n\n### [uvx](https://docs.astral.sh/uv/guides/tools/) 로 빠르게 시작하기\n\nMCP 서버를 실행하는 가장 쉬운 방법은 [uvx를](https://docs.astral.sh/uv/guides/tools/) 사용하는 것입니다.\n\n```sh\nuvx perplexity-advanced-mcp -o # or -p \n```\n\n환경 변수를 사용하여 API 키를 구성할 수도 있습니다.\n\n```sh\nexport OPENROUTER_API_KEY=\"your_key_here\"\n# or\nexport PERPLEXITY_API_KEY=\"your_key_here\"\n\nuvx perplexity-advanced-mcp\n```\n\n메모:\n\n* OpenRouter와 Perplexity API 키를 동시에 제공하면 오류가 발생합니다.\n* CLI 인수와 환경 변수가 모두 제공되는 경우 CLI 인수가 우선합니다.\n\nCLI는 [Typer](https://typer.tiangolo.com/) 로 구축되어 사용자 친화적인 명령줄 환경을 보장합니다.\n\n### MCP 검색 도구\n\n이 패키지에는 `ask_perplexity` 함수를 통해 통합된 MCP 검색 도구가 포함되어 있습니다. 간단한 쿼리와 복잡한 쿼리를 모두 지원하고, 파일 첨부를 처리하여 추가적인 맥락 정보를 제공합니다.\n\n* **간단한 질의:** 빠르고 효율적인 응답을 제공합니다.\n* **복잡한 쿼리:** 세부적인 추론을 수행하고 XML 형식으로 된 파일 첨부를 지원합니다.\n\n## 구성\n\n* **API 키:** 명령줄 옵션이나 환경 변수를 통해 `OPENROUTER_API_KEY` 또는 `PERPLEXITY_API_KEY` 구성합니다.\n* **모델 선택:** 구성( `src/perplexity_advanced_mcp/config.py` )은 쿼리 유형을 특정 모델에 매핑합니다.\n * **[오픈라우터](https://openrouter.ai/) :**\n\n * 간단한 쿼리: `perplexity/sonar`\n * 복잡한 쿼리: `perplexity/sonar-reasoning`\n * **[당혹감](https://docs.perplexity.ai/home) :**\n\n * 간단한 쿼리: `sonar-pro`\n * 복잡한 쿼리: `sonar-reasoning-pro`\n\n## 개발 배경 및 철학\n\n이 프로젝트는 제 개인적인 호기심과 실험 정신에서 시작되었습니다. 최근 [\"바이브 코딩\"](https://x.com/karpathy/status/1886192184808149383) 트렌드를 따라, 코드의 95% 이상을 [Cline](https://cline.bot/) + [Cursor](https://www.cursor.com/) IDE를 통해 작성했습니다. \"말은 싸니까, 코드를 보여줘\"라는 말이 있죠. [Wispr Flow](https://wisprflow.ai/) 의 음성-텍스트 변환 마법 덕분에, 제가 말만 하면 코드가 바로 나타났습니다! 개발의 대부분은 \"xyz 코드를 작성해 줘, 여기 xy z 버그를 수정해 줘.\"와 같은 말과 엔터 키를 누르는 것으로 진행되었습니다. 놀랍게도, 이 모든 기능을 갖춘 프로젝트를 만드는 데 몇 시간도 걸리지 않았습니다.\n\n프로젝트 스캐폴딩부터 파일 구조까지 모든 것이 LLM을 통해 작성되고 검토되었습니다. PyPI 게시를 위한 GitHub Actions 워크플로와 릴리스 승인 프로세스도 Cursor를 통해 처리되었습니다. 실제 개발자로서 제 역할은 다음과 같았습니다.\n\n* AI가 적절한 테스트를 수행할 수 있도록 MCP 서버 시작 및 중지\n* 문제 발생 시 오류 로그 복사 및 제공\n* 인터넷에서 [Python MCP SDK](https://github.com/modelcontextprotocol/python-sdk) 문서 및 예제 찾기 및 제공\n* 올바르지 않은 것으로 보이는 코드에 대한 수정 요청\n\n오늘날 많은 것이 자동화되고 대체될 수 있는 세상에서, 이 MCP가 여러분처럼 MCP를 사용하는 개발자들이 단순한 코드 작성을 넘어 가치를 발견하는 데 도움이 되기를 바랍니다. 이 도구가 여러분이 더 높은 수준의 의사 결정과 고려 사항을 수행할 수 있는 새로운 시대의 개발자로 거듭나는 데 도움이 되기를 바랍니다.\n\n## 개발\n\n이 패키지에 기여하거나 수정하려면:\n\n### 1. **저장소 복제:**\n\n```sh\ngh repo clone code-yeongyu/perplexity-advanced-mcp\n```\n\n### 2. **종속성 설치:**\n\n```sh\nuv sync\n```\n\n### 3. **기여:**\n\n기여를 환영합니다! 기존 코드 스타일과 커밋 지침을 따라주세요.\n\n## 특허\n\n이 프로젝트는 MIT 라이선스에 따라 라이선스가 부여되었습니다.","Korean","ko-KR",{"_36":75,"_40":78,"_45":79},{"_38":76,"_40":41,"_42":77},"Расширенный пакет интеграции, использующий API OpenRouter и Perplexity для предоставления расширенных возможностей веб-поиска, предлагающий как простую, так и сложную обработку запросов с поддержкой прикрепления файлов.","[![Значок оценки безопасности MseeP.ai](https://mseep.net/pr/code-yeongyu-perplexity-advanced-mcp-badge.png)](https://mseep.ai/app/code-yeongyu-perplexity-advanced-mcp)\n\n# Perplexity Advanced MCP\n\n[![GitHub](https://img.shields.io/badge/GitHub-100000?style=for-the-badge\\&logo=github\\&logoColor=white)](https://github.com/code-yeongyu/perplexity-advanced-mcp) [![PyPI](https://img.shields.io/badge/pypi-3775A9?style=for-the-badge\\&logo=pypi\\&logoColor=white)](https://pypi.org/project/perplexity-advanced-mcp) [![значок кузнеца](https://smithery.ai/badge/@code-yeongyu/perplexity-advanced-mcp)](https://smithery.ai/server/@code-yeongyu/perplexity-advanced-mcp)\n\n[한국어](README-ko.md)\n\n***\n\n## Обзор\n\nPerplexity Advanced MCP — это расширенный пакет интеграции, который использует API [OpenRouter](https://openrouter.ai/) и [Perplexity](https://docs.perplexity.ai/home) для предоставления расширенных возможностей обработки запросов. Благодаря интуитивно понятному интерфейсу командной строки и надежному клиенту API этот пакет обеспечивает бесперебойное взаимодействие с моделями ИИ как для простых, так и для сложных запросов.\n\n## Сравнение с [perplexity-mcp](https://github.com/jsonallen/perplexity-mcp)\n\nВ то время как [perplexity-mcp](https://github.com/jsonallen/perplexity-mcp) обеспечивает базовую функциональность веб-поиска с использованием API [Perplexity](https://docs.perplexity.ai/home) AI, Perplexity Advanced MCP предлагает несколько дополнительных функций:\n\n* **Поддержка различных поставщиков:** поддержка API [Perplexity](https://docs.perplexity.ai/home) и [OpenRouter](https://openrouter.ai/) , что дает вам гибкость в выборе поставщика\n* **Оптимизация типа запроса:** различает простые и сложные запросы, оптимизируя затраты и производительность.\n* **Поддержка прикрепленных файлов:** позволяет включать содержимое файлов в контекст ваших запросов, обеспечивая более точные и контекстные ответы.\n* **Улучшенная логика повторных попыток:** реализует надежные механизмы повторных попыток для повышения надежности.\n\nВ целом, это наиболее подходящий MCP для обработки кодовых баз при интеграции с такими редакторами, как [Cline](https://cline.bot/) или [Cursor](https://www.cursor.com/) .\n\n## Функции\n\n* **Унифицированный API-клиент:** поддерживает API [OpenRouter](https://openrouter.ai/) и [Perplexity](https://docs.perplexity.ai/home) с настраиваемыми моделями для обработки простых и сложных запросов.\n* **Интерфейс командной строки (CLI):** управление конфигурацией ключа API и запуск сервера MCP с помощью [Typer](https://typer.tiangolo.com/) .\n* **Расширенная обработка запросов:** включает обработку прикрепленных файлов, что позволяет включать в запросы контекстные данные.\n* **Надежный механизм повторных попыток:** использует Tenacity для логики повторных попыток, чтобы гарантировать последовательную и надежную связь API.\n* **Настраиваемое ведение журнала:** гибкая настройка ведения журнала для детальной отладки и мониторинга выполнения.\n\n## Оптимальная конфигурация ИИ\n\nДля наилучшей работы с помощниками на основе ИИ (например, [Cursor](https://www.cursor.com/) , [Claude for Desktop](https://claude.ai/download) ) я рекомендую добавить следующую конфигурацию в инструкции проекта или правила ИИ:\n\n```xml\n\n \n Perplexity is an LLM that can search the internet, gather information, and answer users' queries.\n\n For example, let's suppose we want to find out the latest version of Python.\n 1. You would search on Google.\n 2. Then read the top two or three results directly to verify.\n\n Perplexity does that work for you.\n\n To answer a user's query, Perplexity searches, opens the top search results, finds information on those websites, and then provides the answer.\n\n Perplexity can be used with two types of queries: simple and complex. Choosing the right query type to fulfill the user's request is most important.\n \n \n \n It's cheap and fast. However, it's not suitable for complex queries. On average, it's more than 10 times cheaper and 3 times faster than complex queries.\n Use it for simple questions such as \"What is the latest version of Python?\"\n \n \n $1/M input tokens\n $1/M output tokens\n \n \n\n \n \n It's slower and more expensive. Compared to simple queries, it's on average more than 10 times more expensive and 3 times slower.\n Use it for more complex requests like \"Analyze the attached code to examine the current status of a specific library and create a migration plan.\"\n \n \n $1/M input tokens\n $5/M output tokens\n \n \n\n \n When reviewing the user's request, if you find anything unexpected, uncertain, or questionable, **and you think you can get answer from the internet**, do not hesitate to use the \"ask_perplexity\" tool to consult Perplexity. However, if the internet is not required to satisfy users' request, it's meaningless to ask to perplexity.\n Since Perplexity is also an LLM, prompt engineering techniques are paramount.\n Remember the basics of prompt engineering, such as providing clear instructions, sufficient context, and examples\n Include as much context and relevant files as possible to smoothly fulfill the user's request. When adding files as attachments, make sure they are absolute paths.\n \n\n```\n\nТакая конфигурация помогает помощникам на базе искусственного интеллекта лучше понимать, когда и как использовать функцию поиска Perplexity, оптимизируя как затраты, так и производительность.\n\n## Использование\n\n### Установка через Smithery\n\nЧтобы автоматически установить Perplexity Advanced MCP для Claude Desktop через [Smithery](https://smithery.ai/server/@code-yeongyu/perplexity-advanced-mcp) :\n\n```bash\nnpx -y @smithery/cli install @code-yeongyu/perplexity-advanced-mcp --client claude\n```\n\n### Быстрый старт с [uvx](https://docs.astral.sh/uv/guides/tools/)\n\nСамый простой способ запустить сервер MCP — использовать [uvx](https://docs.astral.sh/uv/guides/tools/) :\n\n```sh\nuvx perplexity-advanced-mcp -o # or -p \n```\n\nВы также можете настроить ключи API с помощью переменных среды:\n\n```sh\nexport OPENROUTER_API_KEY=\"your_key_here\"\n# or\nexport PERPLEXITY_API_KEY=\"your_key_here\"\n\nuvx perplexity-advanced-mcp\n```\n\nПримечание:\n\n* Предоставление ключей API OpenRouter и Perplexity одновременно приведет к ошибке\n* Если указаны как аргументы CLI, так и переменные среды, аргументы CLI имеют приоритет.\n\nИнтерфейс командной строки создан на основе [Typer](https://typer.tiangolo.com/) , что обеспечивает удобный интерфейс командной строки.\n\n### Инструмент поиска MCP\n\nПакет включает в себя поисковый инструмент MCP, интегрированный через функцию `ask_perplexity` . Он поддерживает как простые, так и сложные запросы и обрабатывает прикрепленные файлы для предоставления дополнительного контекста.\n\n* **Простые запросы:** обеспечивают быстрые и эффективные ответы.\n* **Сложные запросы:** выполняет подробные рассуждения и поддерживает вложения файлов в формате XML.\n\n## Конфигурация\n\n* **Ключи API:** настройте `OPENROUTER_API_KEY` или `PERPLEXITY_API_KEY` с помощью параметров командной строки или переменных среды.\n* **Выбор модели:** конфигурация (в `src/perplexity_advanced_mcp/config.py` ) сопоставляет типы запросов с определенными моделями:\n * **[OpenRouter](https://openrouter.ai/) :**\n\n * Простые запросы: `perplexity/sonar`\n * Сложные запросы: `perplexity/sonar-reasoning`\n * **[Недоумение](https://docs.perplexity.ai/home) :**\n\n * Простые запросы: `sonar-pro`\n * Сложные запросы: `sonar-reasoning-pro`\n\n## Предыстория и философия развития\n\nЭтот проект возник из моего личного любопытства и экспериментов. Следуя недавней тенденции [\"vibe coding\"](https://x.com/karpathy/status/1886192184808149383) , более 95% кода было написано с помощью [Cline](https://cline.bot/) + [Cursor](https://www.cursor.com/) IDE. Говорят, \"разговоры ничего не стоят, покажите мне код\" - ну, с помощью волшебства речи в текст [Wispr Flow](https://wisprflow.ai/) я буквально просто говорил, и код появлялся! Большая часть разработки была сделана мной, когда я говорил что-то вроде \"Напишите мне код для xyz, исправьте ошибку здесь xy z\" и нажимал Enter. Примечательно, что создание этого полностью функционального проекта заняло менее нескольких часов.\n\nОт проектирования до структуры файлов, все было написано и проверено через LLM. Даже рабочий процесс GitHub Actions для публикации PyPI и процесс утверждения релиза обрабатывались через Cursor. Как разработчик-человек, моя роль заключалась в следующем:\n\n* Запуск и остановка сервера MCP для помощи ИИ в проведении надлежащего тестирования\n* Копирование и предоставление журналов ошибок при возникновении проблем\n* Поиск и предоставление документации и примеров [Python MCP SDK](https://github.com/modelcontextprotocol/python-sdk) из Интернета\n* Запрос на внесение изменений в код, который показался вам некорректным\n\nВ современном мире, где многие вещи можно автоматизировать и заменить, я надеюсь, что этот MCP поможет разработчикам, таким как вы, которые используют его, обнаружить ценность, выходящую за рамки простого написания кода. Пусть этот инструмент поможет вам стать разработчиком новой эры, который может принимать решения и соображения более высокого уровня.\n\n## Разработка\n\nЧтобы внести свой вклад или изменить этот пакет:\n\n### 1. **Клонируйте репозиторий:**\n\n```sh\ngh repo clone code-yeongyu/perplexity-advanced-mcp\n```\n\n### 2. **Установите зависимости:**\n\n```sh\nuv sync\n```\n\n### 3. **Внесите свой вклад:**\n\nВклады приветствуются! Пожалуйста, следуйте существующему стилю кода и рекомендациям по коммитам.\n\n## Лицензия\n\nДанный проект лицензирован по лицензии MIT.","Russian","ru-RU",{"_36":81,"_40":85,"_45":86},{"_38":82,"_40":83,"_42":84},"这是一个先进的集成包，利用 OpenRouter 和 Perplexity API 来提供增强的网络搜索功能，提供简单和复杂的查询处理以及文件附件支持。","Perplexity 高级 MCP","# Perplexity 高级 MCP\n\n[![GitHub](https://img.shields.io/badge/GitHub-100000?style=for-the-badge\\&logo=github\\&logoColor=white)](https://github.com/code-yeongyu/perplexity-advanced-mcp) [![PyPI](https://img.shields.io/badge/pypi-3775A9?style=for-the-badge\\&logo=pypi\\&logoColor=white)](https://pypi.org/project/perplexity-advanced-mcp) [![铁匠徽章](https://smithery.ai/badge/@code-yeongyu/perplexity-advanced-mcp)](https://smithery.ai/server/@code-yeongyu/perplexity-advanced-mcp)\n\n[韩国](README-ko.md)\n\n***\n\n## 概述\n\nPerplexity Advanced MCP 是一个高级集成包，它利用[OpenRouter](https://openrouter.ai/)和[Perplexity](https://docs.perplexity.ai/home) API 提供增强的查询处理功能。该包拥有直观的命令行界面和强大的 API 客户端，可与 AI 模型无缝交互，处理简单和复杂的查询。\n\n## 与[perplexity-mcp 的](https://github.com/jsonallen/perplexity-mcp)比较\n\n[perplexity-mcp](https://github.com/jsonallen/perplexity-mcp)使用[Perplexity](https://docs.perplexity.ai/home) AI 的 API 提供基本的网页搜索功能，而 Perplexity Advanced MCP 则提供了一些附加功能：\n\n* \\*\\*多供应商支持：\\*\\*支持[Perplexity](https://docs.perplexity.ai/home)和[OpenRouter](https://openrouter.ai/) API，让您可以灵活地选择供应商\n* \\*\\*查询类型优化：\\*\\*区分简单查询和复杂查询，优化成本和性能\n* \\*\\*文件附件支持：\\*\\*允许将文件内容作为上下文包含在查询中，从而实现更精确、更具上下文的响应\n* \\*\\*增强的重试逻辑：\\*\\*实施强大的重试机制以提高可靠性\n\n总体而言，当与[Cline](https://cline.bot/)或[Cursor](https://www.cursor.com/)等编辑器集成时，这是最适合处理代码库的 MCP。\n\n## 特征\n\n* \\*\\*统一 API 客户端：\\*\\*支持[OpenRouter](https://openrouter.ai/)和[Perplexity](https://docs.perplexity.ai/home) API，并具有可配置模型来处理简单和复杂的查询。\n* \\*\\*命令行界面（CLI）：\\*\\*管理 API 密钥配置并使用[Typer](https://typer.tiangolo.com/)运行 MCP 服务器。\n* \\*\\*高级查询处理：\\*\\*结合文件附件处理，允许您在查询中包含上下文数据。\n* \\*\\*强大的重试机制：\\*\\*利用 Tenacity 进行重试逻辑，确保 API 通信一致、可靠。\n* \\*\\*可定制的日志记录：\\*\\*灵活的日志记录配置，用于详细的调试和运行时监控。\n\n## 最佳AI配置\n\n为了获得 AI 助手（例如[Cursor](https://www.cursor.com/) 、 [Claude for Desktop](https://claude.ai/download) ）的最佳体验，我建议在项目说明或 AI 规则中添加以下配置：\n\n```xml\n\n \n Perplexity is an LLM that can search the internet, gather information, and answer users' queries.\n\n For example, let's suppose we want to find out the latest version of Python.\n 1. You would search on Google.\n 2. Then read the top two or three results directly to verify.\n\n Perplexity does that work for you.\n\n To answer a user's query, Perplexity searches, opens the top search results, finds information on those websites, and then provides the answer.\n\n Perplexity can be used with two types of queries: simple and complex. Choosing the right query type to fulfill the user's request is most important.\n \n \n \n It's cheap and fast. However, it's not suitable for complex queries. On average, it's more than 10 times cheaper and 3 times faster than complex queries.\n Use it for simple questions such as \"What is the latest version of Python?\"\n \n \n $1/M input tokens\n $1/M output tokens\n \n \n\n \n \n It's slower and more expensive. Compared to simple queries, it's on average more than 10 times more expensive and 3 times slower.\n Use it for more complex requests like \"Analyze the attached code to examine the current status of a specific library and create a migration plan.\"\n \n \n $1/M input tokens\n $5/M output tokens\n \n \n\n \n When reviewing the user's request, if you find anything unexpected, uncertain, or questionable, **and you think you can get answer from the internet**, do not hesitate to use the \"ask_perplexity\" tool to consult Perplexity. However, if the internet is not required to satisfy users' request, it's meaningless to ask to perplexity.\n Since Perplexity is also an LLM, prompt engineering techniques are paramount.\n Remember the basics of prompt engineering, such as providing clear instructions, sufficient context, and examples\n Include as much context and relevant files as possible to smoothly fulfill the user's request. When adding files as attachments, make sure they are absolute paths.\n \n\n```\n\n此配置可帮助 AI 助手更好地了解何时以及如何使用 Perplexity 搜索功能，从而优化成本和性能。\n\n## 用法\n\n### 通过 Smithery 安装\n\n要通过[Smithery](https://smithery.ai/server/@code-yeongyu/perplexity-advanced-mcp)自动为 Claude Desktop 安装 Perplexity Advanced MCP：\n\n```bash\nnpx -y @smithery/cli install @code-yeongyu/perplexity-advanced-mcp --client claude\n```\n\n### [uvx](https://docs.astral.sh/uv/guides/tools/)快速入门\n\n运行 MCP 服务器最简单的方法是使用[uvx](https://docs.astral.sh/uv/guides/tools/) ：\n\n```sh\nuvx perplexity-advanced-mcp -o # or -p \n```\n\n您还可以使用环境变量配置 API 密钥：\n\n```sh\nexport OPENROUTER_API_KEY=\"your_key_here\"\n# or\nexport PERPLEXITY_API_KEY=\"your_key_here\"\n\nuvx perplexity-advanced-mcp\n```\n\n笔记：\n\n* 同时提供 OpenRouter 和 Perplexity API 密钥将导致错误\n* 当同时提供 CLI 参数和环境变量时，CLI 参数优先\n\nCLI 使用[Typer](https://typer.tiangolo.com/)构建，确保用户友好的命令行体验。\n\n### MCP 搜索工具\n\n该软件包包含一个通过`ask_perplexity`函数集成的 MCP 搜索工具。它支持简单和复杂的查询，并可处理文件附件以提供额外的上下文信息。\n\n* \\*\\*简单查询：\\*\\*提供快速、高效的响应。\n* \\*\\*复杂查询：\\*\\*进行详细推理并支持 XML 格式的文件附件。\n\n## 配置\n\n* \\*\\*API 密钥：\\*\\*通过命令行选项或环境变量配置`OPENROUTER_API_KEY`或`PERPLEXITY_API_KEY` 。\n* \\*\\*模型选择：\\*\\*配置（在`src/perplexity_advanced_mcp/config.py`中）将查询类型映射到特定模型：\n * **[OpenRouter](https://openrouter.ai/) ：**\n\n * 简单查询： `perplexity/sonar`\n * 复杂查询： `perplexity/sonar-reasoning`\n * **[困惑](https://docs.perplexity.ai/home)：**\n\n * 简单查询： `sonar-pro`\n * 复杂查询： `sonar-reasoning-pro`\n\n## 开发背景与理念\n\n这个项目源于我个人的好奇心和实验。顺应最近的[“氛围编码”](https://x.com/karpathy/status/1886192184808149383)潮流，超过 95% 的代码都是通过[Cline](https://cline.bot/) + [Cursor](https://www.cursor.com/) IDE 编写的。俗话说“说话不费力，代码见分晓”——有了[Wispr Flow](https://wisprflow.ai/)的语音转文本功能，我只需开口说话，代码就能出来！大部分开发工作都是我通过说“帮我写 xyz 的代码，修复 xyz 的错误”之类的指令，然后按下回车键完成的。值得一提的是，创建这个功能齐全的项目只花了不到几个小时。\n\n从项目脚手架到文件结构，所有内容均通过 LLM 编写和审核。甚至 PyPI 发布的 GitHub Actions 工作流和发布审批流程也通过 Cursor 处理。作为一名人类开发人员，我的职责是：\n\n* 启动和停止 MCP 服务器以帮助 AI 进行正确的测试\n* 出现问题时复制并提供错误日志\n* 从互联网上查找并提供[Python MCP SDK](https://github.com/modelcontextprotocol/python-sdk)文档和示例\n* 请求修改看起来不正确的代码\n\n在当今世界，许多事物都可以自动化和替代，我希望这款 MCP 能够帮助像您一样使用它的开发者发现超越编写代码的价值。愿它助您成为能够做出更高层次决策和考量的新一代开发者。\n\n## 发展\n\n要贡献或修改此包：\n\n### 1.**克隆存储库：**\n\n```sh\ngh repo clone code-yeongyu/perplexity-advanced-mcp\n```\n\n### 2.**安装依赖项：**\n\n```sh\nuv sync\n```\n\n### 3.**贡献：**\n\n欢迎贡献代码！请遵循现有的代码风格和提交指南。\n\n## 执照\n\n该项目已获得 MIT 许可。","Chinese","zh-CN","githubUser","locale",{"_90":91,"_92":93,"_20":111,"_120":49,"_121":49,"_122":123,"_126":127,"_147":-5,"_23":148,"_149":150,"_151":152,"_40":41,"_184":140,"_185":-5,"_186":-5,"_187":-5,"_188":-5,"_189":190,"_196":-5,"_159":136,"_197":198,"_202":203,"_8":231,"_232":233},"addedAt","2025-03-21T11:17:57.909909890Z","argumentsJsonSchema",{"_94":95,"_103":104,"_12":110},"oneOf",[96,100],{"_97":98},"required",[99],"OPENROUTER_API_KEY",{"_97":101},[102],"PERPLEXITY_API_KEY","properties",{"_99":105,"_102":108},{"_38":106,"_12":107},"Your OpenRouter API key","string",{"_38":109,"_12":107},"Your Perplexity API key","object",[112,113,114,115,116,117,118,119],"category:search","category:browser-automation","author:claimed","hosting:remote-capable","language:python","security_grade:a","quality_grade:a","license_grade:f","descriptionMarkdown","descriptionPlainText","githubProject",{"_42":50,"_124":125},"url","https://github.com/code-yeongyu/perplexity-advanced-mcp","githubRepository",{"_128":129,"_130":131,"_132":133,"_134":135,"_40":136,"_137":138,"_142":143,"_144":-5,"_145":146},"defaultBranch","master","fullName","code-yeongyu/perplexity-advanced-mcp","githubTreeId",115800,"language","Python","perplexity-advanced-mcp","owner",{"_139":140,"_8":141},"login","code-yeongyu","eW25BmCIHe","recentStargazers",1,"spdxLicense","stargazers",22,"glamaJson",2370,"inspectable",true,"integrations",[153,163,170,177],{"_154":155,"_38":161,"_8":162},"brand",{"_156":157,"_40":158,"_159":160},"logoSvgPath","M12 .297c-6.63 0-12 5.373-12 12 0 5.303 3.438 9.8 8.205 11.385.6.113.82-.258.82-.577 0-.285-.01-1.04-.015-2.04-3.338.724-4.042-1.61-4.042-1.61C4.422 18.07 3.633 17.7 3.633 17.7c-1.087-.744.084-.729.084-.729 1.205.084 1.838 1.236 1.838 1.236 1.07 1.835 2.809 1.305 3.495.998.108-.776.417-1.305.76-1.605-2.665-.3-5.466-1.332-5.466-5.93 0-1.31.465-2.38 1.235-3.22-.135-.303-.54-1.523.105-3.176 0 0 1.005-.322 3.3 1.23.96-.267 1.98-.399 3-.405 1.02.006 2.04.138 3 .405 2.28-1.552 3.285-1.23 3.285-1.23.645 1.653.24 2.873.12 3.176.765.84 1.23 1.91 1.23 3.22 0 4.61-2.805 5.625-5.475 5.92.42.36.81 1.096.81 2.22 0 1.606-.015 2.896-.015 3.286 0 .315.21.69.825.57C20.565 22.092 24 17.592 24 12.297c0-6.627-5.373-12-12-12","GitHub","slug","github","Provides integration with GitHub for source code hosting and version control.","t57sdw9k8y",{"_154":164,"_38":168,"_8":169},{"_156":165,"_40":166,"_159":167},"M22.3977 7.0896h-2.3106V.0676l-7.5094 6.3542V.1577h-1.1554v6.1966L4.4904 0v7.0896H1.6023v10.3976h2.8882V24l6.932-6.3591v6.2005h1.1554v-6.0469l6.9318 6.1807v-6.4879h2.8882V7.0896zm-3.4657-4.531v4.531h-5.355l5.355-4.531zm-13.2862.0676 4.8691 4.4634H5.6458V2.6262zM2.7576 16.332V8.245h7.8476l-6.1149 6.1147v1.9723H2.7576zm2.8882 5.0404v-3.8852h.0001v-2.6488l5.7763-5.7764v7.0111l-5.7764 5.2993zm12.7086.0248-5.7766-5.1509V9.0618l5.7766 5.7766v6.5588zm2.8882-5.0652h-1.733v-1.9723L13.3948 8.245h7.8478v8.087z","Perplexity","perplexity","Provides a tool for asking internet search queries through Perplexity AI, supporting both simple and complex queries with file attachment capabilities. Optimized for different query types to balance cost and performance.","2cn99jeh3l",{"_154":171,"_38":175,"_8":176},{"_156":172,"_40":173,"_159":174},"M23.922 13.58v3.912L20.55 18.72l-.078.055.052.037 3.45-1.256.026-.036v-3.997l-.053-.036-.025.092z M23.621 5.618l-3.04 1.107v3.912l3.339-1.215V5.509zM23.92 13.457V9.544l-3.336 1.215v3.913zM20.47 14.71V10.8L17.17 12v3.913zM17.034 19.996v-3.912l-3.313 1.206v3.912zM17.17 16.057v3.868l3.314-1.206V14.85l-3.314 1.206zm2.093 1.882c-.367.134-.663-.074-.663-.463s.296-.814.663-.947c.365-.133.662.075.662.464s-.297.814-.662.946z M13.225 9.315l.365-.132-3.285-1.197-3.323 1.21.102.037 3.184 1.16zM20.507 10.664V6.751L17.17 7.965v3.913zM17.058 11.918V8.005l-3.302 1.202v3.912zM13.643 9.246l-3.336 1.215v3.913l3.336-1.215zM6.907 13.165l3.322 1.209v-3.913L6.907 9.252z M10.34 7.873l3.281 1.193V5.198l-3.28-1.193zM20.507 2.715L17.19 3.922v3.913l3.317-1.207zM16.95 3.903L13.724 2.73l-3.269 1.19 3.225 1.174zM15.365 4.606l-1.624.592v3.868l3.317-1.207V3.991l-1.693.615zm-.391 2.778c-.367.134-.662-.074-.662-.464s.295-.813.662-.946c.366-.133.663.074.663.464s-.297.813-.663.946z M10.229 18.41v-3.914l-3.322-1.209V17.2zM13.678 17.182v-3.913l-3.371 1.227v3.913z M13.756 17.154l3.3-1.2V12.04l-3.3 1.2zM13.678 21.217l-3.371 1.227v-3.912h-.078v3.912l-3.322-1.209v-3.913l-.053-.058-.025-.06-3.336-1.21v-3.948l.034.013 3.287 1.196.015-.078-3.261-1.187 3.26-1.187v-.109L3.876 9.62l-.307-.112 3.26-1.188v.877l.079-.055V6.769l3.257 1.185.058-.061L7.084 6.75l-.102-.037 3.24-1.179v-.083L6.854 6.677v.018l-.025.018v1.523L3.44 9.47v.02l-.025.017v4.007l-3.39 1.233v.019L0 14.784v3.995l.025.037 3.4 1.237.008-.006.007.01 3.4 1.238.008-.006.006.01 3.4 1.237.014-.009.012.01 3.45-1.256.026-.037-.078-.027zM3.493 9.563l3.257 1.185-3.257 1.187V9.562zM3.4 19.96L.078 18.752v-3.913l2.361.86.96.349v3.913zm.015-3.99L.335 14.85l-.182-.066 3.262-1.187v2.374zm3.399 5.231l-3.321-1.209v-3.912l3.321 1.209v3.912zM23.791 5.434l-3.21-1.17v2.338zM20.387 2.643l-3.24-1.18-3.27 1.19 3.247 1.182z","PyPI","pypi","Enables distribution of the package through the Python Package Index, allowing users to install the MCP via pip or uvx.","dc3dyquaaa",{"_154":178,"_38":182,"_8":183},{"_156":179,"_40":180,"_159":181},"M21.818 3.646H2.182C.982 3.646 0 4.483 0 5.505v3.707h2.182V5.486h19.636v13.036H2.182v-3.735H0v3.726c0 1.022.982 1.84 2.182 1.84h19.636c1.2 0 2.182-.818 2.182-1.84V5.505c0-1.032-.982-1.859-2.182-1.859Zm-10.909 12.07L15.273 12l-4.364-3.717v2.787H0v1.859h10.909v2.787Z","Typer","typer","Implements a command-line interface using Typer for managing API key configuration and running the MCP server.","w3iqwiiqjj","namespace","npmPackage","prompts","redditPostForeignId","resources","scores",{"_191":192,"_193":194,"_195":194},"license",0,"quality",100,"security","simpleIcon","supportedPlatforms",[199,200,201],"WINDOWS","LINUX","MACOS","tools",[204],{"_40":205,"_38":206,"_207":208},"ask_perplexity","Perplexity is fundamentally an LLM that can search the internet, gather information, and answer users' queries.\n\n For example, let's suppose we want to find out the latest version of Python.\n 1. You would search on Google.\n 2. Then read the top two or three results directly to verify.\n\n Perplexity does that work for you.\n\n To answer a user's query, Perplexity searches, opens the top search results, finds information on those websites, and then provides the answer.\n\n Perplexity can be used with two types of queries: simple and complex. Choosing the right query type to fulfill the user's request is most important.\n\n SIMPLE Query:\n - Cheap and fast (on average, 10x cheaper and 3x faster than complex queries).\n - Suitable for straightforward questions such as \"What is the latest version of Python?\"\n - Pricing: $1/M input tokens, $1/M output tokens.\n\n COMPLEX Query:\n - Slower and more expensive (on average, 10x more expensive and 3x slower).\n - Suitable for tasks requiring multiple steps of reasoning or deep analysis, such as \"Analyze the attached code to examine the current status of a specific library and create a migration plan.\"\n - Pricing: $1/M input tokens, $5/M output tokens.\n\n Instructions:\n - When reviewing the user's request, if you find anything unexpected, uncertain, or questionable, do not hesitate to use the \"ask_perplexity\" tool to consult Perplexity.\n - Since Perplexity is also an LLM, prompt engineering techniques are paramount.\n - Remember the basics of prompt engineering, such as providing clear instructions, sufficient context, and examples.\n - Include as much context and relevant files as possible to smoothly fulfill the user's request.\n - IMPORTANT: When adding files as attachments, you MUST use absolute paths (e.g., '/absolute/path/to/file.py'). Relative paths will not work.\n\n Note: All queries must be in English for optimal results.\n ","inputSchema",{"_12":110,"_10":209,"_97":210,"_103":214},"ask_perplexityArguments",[211,212,213],"query","query_type","attachment_paths",{"_211":215,"_212":218,"_213":225},{"_12":107,"_10":216,"_38":217},"Query","The query to search for",{"_219":220,"_12":107,"_10":223,"_38":224},"enum",[221,222],"simple","complex","Query Type","Type of query to determine model selection",{"_12":226,"_227":228,"_10":229,"_38":230},"array","items",{"_12":107},"Attachment Paths","An optional list of absolute file paths to attach as context for the search query","sok882f31z","updatedAt","2025-05-07T14:00:32.838838674Z","meta",[236,238,240,247,250,252,255,258,261,264,267,269,272,275,277,279,281,283,285],{"_10":237},"Perplexity Advanced MCP | Glama",{"_239":49,"_40":38},"content",{"_241":242,"_243":244,"_245":246},"href","https://glama.ai/mcp/servers/@code-yeongyu/perplexity-advanced-mcp","rel","canonical","tagName","link",{"_239":41,"_248":249},"property","og:title",{"_239":49,"_248":251},"og:description",{"_239":253,"_248":254},"https://glama.ai/logo.png","og:logo",{"_239":256,"_248":257},"https://glama.ai/mcp/servers/@code-yeongyu/perplexity-advanced-mcp/og-image","og:image",{"_239":259,"_248":260},"630","og:image:height",{"_239":262,"_248":263},"1200","og:image:width",{"_239":265,"_248":266},"website","og:type",{"_239":242,"_248":268},"og:url",{"_239":270,"_248":271},"Glama – MCP Hosting Platform","og:site_name",{"_239":273,"_40":274},"summary_large_image","twitter:card",{"_239":270,"_40":276},"twitter:title",{"_239":49,"_40":278},"twitter:description",{"_239":256,"_40":280},"twitter:image",{"_239":259,"_40":282},"twitter:image:height",{"_239":262,"_40":284},"twitter:image:width",{"_286":287},"script:ld+json",{"_288":289,"_290":291},"@context","http://schema.org","@graph",[292],{"_288":293,"_294":295,"_296":297},"https://schema.org","@type","BreadcrumbList","itemListElement",[298,304,308,312],{"_294":299,"_300":301,"_40":302,"_303":143},"ListItem","item","https://glama.ai/","Glama","position",{"_294":299,"_300":305,"_40":306,"_303":307},"https://glama.ai/mcp","MCP",2,{"_294":299,"_300":309,"_40":310,"_303":311},"https://glama.ai/mcp/servers","Servers",3,{"_294":299,"_300":242,"_40":41,"_303":313},4,"visitorIsMaintainer",false,"routes/_public/mcp/servers/~namespace/~slug/_pages/reviews/_route",{"_87":-5,"_36":318,"_234":353,"_388":-5},{"_90":91,"_92":319,"_20":328,"_120":49,"_121":49,"_122":329,"_126":330,"_147":-5,"_23":148,"_149":150,"_151":332,"_40":41,"_184":140,"_185":-5,"_186":-5,"_187":-5,"_188":-5,"_189":341,"_196":-5,"_159":136,"_197":342,"_202":343,"_8":231,"_232":233},{"_94":320,"_103":325,"_12":110},[321,323],{"_97":322},[99],{"_97":324},[102],{"_99":326,"_102":327},{"_38":106,"_12":107},{"_38":109,"_12":107},[112,113,114,115,116,117,118,119],{"_42":50,"_124":125},{"_128":129,"_130":131,"_132":133,"_134":135,"_40":136,"_137":331,"_142":143,"_144":-5,"_145":146},{"_139":140,"_8":141},[333,335,337,339],{"_154":334,"_38":161,"_8":162},{"_156":157,"_40":158,"_159":160},{"_154":336,"_38":168,"_8":169},{"_156":165,"_40":166,"_159":167},{"_154":338,"_38":175,"_8":176},{"_156":172,"_40":173,"_159":174},{"_154":340,"_38":182,"_8":183},{"_156":179,"_40":180,"_159":181},{"_191":192,"_193":194,"_195":194},[199,200,201],[344],{"_40":205,"_38":206,"_207":345},{"_12":110,"_10":209,"_97":346,"_103":347},[211,212,213],{"_211":348,"_212":349,"_213":351},{"_12":107,"_10":216,"_38":217},{"_219":350,"_12":107,"_10":223,"_38":224},[221,222],{"_12":226,"_227":352,"_10":229,"_38":230},{"_12":107},[354,356,358,360,362,363,364,365,366,367,368,369,370,371,372,373,374,375,376],{"_10":355},"Reviews | Perplexity Advanced MCP | Glama",{"_239":357,"_40":38},"Read community reviews and feedback for Perplexity Advanced MCP. Discover user experiences, ratings, strengths, and potential limitations of this MCP implementation from actual users.",{"_241":359,"_243":244,"_245":246},"https://glama.ai/mcp/servers/@code-yeongyu/perplexity-advanced-mcp/reviews",{"_239":361,"_248":249},"Reviews | Perplexity Advanced MCP",{"_239":357,"_248":251},{"_239":253,"_248":254},{"_239":256,"_248":257},{"_239":259,"_248":260},{"_239":262,"_248":263},{"_239":265,"_248":266},{"_239":359,"_248":268},{"_239":270,"_248":271},{"_239":273,"_40":274},{"_239":270,"_40":276},{"_239":357,"_40":278},{"_239":256,"_40":280},{"_239":259,"_40":282},{"_239":262,"_40":284},{"_286":377},{"_288":289,"_290":378},[379],{"_288":293,"_294":295,"_296":380},[381,382,383,384,385],{"_294":299,"_300":301,"_40":302,"_303":143},{"_294":299,"_300":305,"_40":306,"_303":307},{"_294":299,"_300":309,"_40":310,"_303":311},{"_294":299,"_300":242,"_40":41,"_303":313},{"_294":299,"_300":359,"_40":386,"_303":387},"Reviews",5,"review","actionData","errors"]

Rate & Review

MCP directory API