[{"_1":2,"_284":-5,"_285":-5},"loaderData",{"_3":4,"_22":23,"_230":231},"root",{"_5":6,"_7":8},"toasts",[],"visitor",{"_9":10,"_11":12},"viewportWidth",1280,"visitorSession",{"_13":14,"_16":17,"_18":-5,"_19":20,"_21":-5},"attributes",[15],"bot","id",14674407,"membership","referenceId","0199368b-156e-7599-a55d-ca69a5c1bb1e","userAccount","routes/_public/mcp/servers/~namespace/~slug/_pages/_layout",{"_24":25,"_66":-5,"_67":26,"_27":68,"_146":147,"_229":133},"availableLocales",[26,38,45,52,59],{"_27":28,"_31":35,"_36":37},"mcpServer",{"_29":30,"_31":32,"_33":34},"description","An MCP server that uses Groq's API to expose raw chain-of-thought tokens from Qwen's qwq model, enabling LLMs to think step-by-step before responding.","name","Chain of Thought MCP Server","readme","# Chain of Thought MCP Server\n\nAnthropic's recent article [\"The \"think\" tool: Enabling Claude to stop and think in complex tool use situations\"](https://www.anthropic.com/engineering/claude-think-tool) shows that using an external `think` tool notably increases performance on [SWE Bench](https://www.swebench.com/).\n\nThis MCP Server uses Groq's API to call LLMs which expose raw chain-of-thought tokens from Qwen's [qwq model](https://qwenlm.github.io/blog/qwq-32b/).\n\n## Installation \n\n1. Clone this repository to your local machine.\n2. Run ```uv sync``` to install depencies\n3. Get a Groq API key from [here](https://console.groq.com/keys).\n4. Update your mcp configuration with:\n\n```json\n\"mcpServers\": {\n \"chain_of_thought\": {\n \"command\": \"uv\",\n \"args\": [\n \"--directory\",\n \"path/to/cot-mcp-server\",\n \"run\",\n \"src/server.py\"\n ],\n \"env\": {\n \"GROQ_API_KEY\": \"your-groq-api-key\"\n }\n }\n}\n```\n\nThe path should be the local path to this repository. You can get this easily by running `pwd` in the terminal from the root of the repository.\n\n## Instructing The AI To Use This MCP Server\n\nI personally prefer the agent call this tool on every request to increase performance. I add this to my rules for the agent:\n\n```xml\n\n\nYou should call the mcp chain_of_thought tool every time you talk to the user, which generates a chain-of-thought stream which you will use to complete the user's request.\n\n\nBefore taking any action or responding to the user use the chain of thought tool as a scratchpad to:\n- List the specific rules that apply to the current request\n- Check if all required information is collected\n- Verify that the planned action complies with all policies\n- Iterate over tool results for correctness \n\nHere are some examples of what to iterate over inside the think tool:\n\nUser wants to cancel flight ABC123\n- Need to verify: user ID, reservation ID, reason\n- Check cancellation rules:\n * Is it within 24h of booking?\n * If not, check ticket class and insurance\n- Verify no segments flown or are in the past\n- Plan: collect missing info, verify rules, get confirmation\n\n\n\nUser wants to book 3 tickets to NYC with 2 checked bags each\n- Need user ID to check:\n * Membership tier for baggage allowance\n * Which payments methods exist in profile\n- Baggage calculation:\n * Economy class × 3 passengers\n * If regular member: 1 free bag each → 3 extra bags = $150\n * If silver member: 2 free bags each → 0 extra bags = $0\n * If gold member: 3 free bags each → 0 extra bags = $0\n- Payment rules to verify:\n * Max 1 travel certificate, 1 credit card, 3 gift cards\n * All payment methods must be in profile\n * Travel certificate remainder goes to waste\n- Plan:\n1. Get user ID\n2. Verify membership level for bag fees\n3. Check which payment methods in profile and if their combination is allowed\n4. Calculate total: ticket price + any bag fees\n5. Get explicit confirmation for booking\n\n\n\n```","English","tag","en-US",{"_27":39,"_31":43,"_36":44},{"_29":40,"_31":41,"_33":42},"Un servidor MCP que utiliza la API de Groq para exponer tokens de cadena de pensamiento sin procesar del modelo qwq de Qwen, lo que permite a los LLM pensar paso a paso antes de responder.","Servidor MCP de cadena de pensamiento","# Servidor MCP de cadena de pensamiento\n\nEl artículo reciente de Anthropic [\"La herramienta \"pensar\": permitir que Claude se detenga y piense en situaciones complejas de uso de herramientas\"](https://www.anthropic.com/engineering/claude-think-tool) muestra que el uso de una herramienta `think` externa aumenta notablemente el rendimiento en [SWE Bench](https://www.swebench.com/) .\n\nEste servidor MCP utiliza la API de Groq para llamar a los LLM que exponen tokens de cadena de pensamiento sin procesar del [modelo qwq](https://qwenlm.github.io/blog/qwq-32b/) de Qwen.\n\n## Instalación\n\n1. Clone este repositorio en su máquina local.\n2. Ejecute `uv sync` para instalar las dependencias\n3. Obtenga una clave API de Groq desde [aquí](https://console.groq.com/keys) .\n4. Actualice su configuración de mcp con:\n\n```json\n\"mcpServers\": {\n \"chain_of_thought\": {\n \"command\": \"uv\",\n \"args\": [\n \"--directory\",\n \"path/to/cot-mcp-server\",\n \"run\",\n \"src/server.py\"\n ],\n \"env\": {\n \"GROQ_API_KEY\": \"your-groq-api-key\"\n }\n }\n}\n```\n\nLa ruta debe ser la ruta local de este repositorio. Puede obtenerla fácilmente ejecutando `pwd` en la terminal desde la raíz del repositorio.\n\n## Dar instrucciones a la IA para que utilice este servidor MCP\n\nPersonalmente, prefiero que el agente llame a esta herramienta en cada solicitud para mejorar el rendimiento. Añado esto a mis reglas para el agente:\n\n```xml\n\n\nYou should call the mcp chain_of_thought tool every time you talk to the user, which generates a chain-of-thought stream which you will use to complete the user's request.\n\n\nBefore taking any action or responding to the user use the chain of thought tool as a scratchpad to:\n- List the specific rules that apply to the current request\n- Check if all required information is collected\n- Verify that the planned action complies with all policies\n- Iterate over tool results for correctness \n\nHere are some examples of what to iterate over inside the think tool:\n\nUser wants to cancel flight ABC123\n- Need to verify: user ID, reservation ID, reason\n- Check cancellation rules:\n * Is it within 24h of booking?\n * If not, check ticket class and insurance\n- Verify no segments flown or are in the past\n- Plan: collect missing info, verify rules, get confirmation\n\n\n\nUser wants to book 3 tickets to NYC with 2 checked bags each\n- Need user ID to check:\n * Membership tier for baggage allowance\n * Which payments methods exist in profile\n- Baggage calculation:\n * Economy class × 3 passengers\n * If regular member: 1 free bag each → 3 extra bags = $150\n * If silver member: 2 free bags each → 0 extra bags = $0\n * If gold member: 3 free bags each → 0 extra bags = $0\n- Payment rules to verify:\n * Max 1 travel certificate, 1 credit card, 3 gift cards\n * All payment methods must be in profile\n * Travel certificate remainder goes to waste\n- Plan:\n1. Get user ID\n2. Verify membership level for bag fees\n3. Check which payment methods in profile and if their combination is allowed\n4. Calculate total: ticket price + any bag fees\n5. Get explicit confirmation for booking\n\n\n\n```","Spanish","es-ES",{"_27":46,"_31":50,"_36":51},{"_29":47,"_31":48,"_33":49},"Groq の API を使用して Qwen の qwq モデルから生の思考連鎖トークンを公開し、LLM が応答する前にステップごとに考えることができる MCP サーバー。","Chain of Thought MCP サーバー","# Chain of Thought MCP サーバー\n\nAnthropic の最近の記事[「考える」ツール: 複雑なツール使用状況で Claude が立ち止まって考えることを可能にする」では、](https://www.anthropic.com/engineering/claude-think-tool)外部の`think`ツールを使用すると[SWE Bench](https://www.swebench.com/)のパフォーマンスが著しく向上することが示されています。\n\nこの MCP サーバーは、Groq の API を使用して、Qwen の[qwq モデル](https://qwenlm.github.io/blog/qwq-32b/)から生の思考連鎖トークンを公開する LLM を呼び出します。\n\n## インストール\n\n1. このリポジトリをローカルマシンにクローンします。\n2. 依存関係をインストールするには、 `uv sync`を実行します。\n3. [ここ](https://console.groq.com/keys)から Groq API キーを取得します。\n4. 次のように mcp 構成を更新します。\n\n```json\n\"mcpServers\": {\n \"chain_of_thought\": {\n \"command\": \"uv\",\n \"args\": [\n \"--directory\",\n \"path/to/cot-mcp-server\",\n \"run\",\n \"src/server.py\"\n ],\n \"env\": {\n \"GROQ_API_KEY\": \"your-groq-api-key\"\n }\n }\n}\n```\n\nパスはこのリポジトリへのローカルパスである必要があります。リポジトリのルートからターミナルで`pwd`を実行すると簡単に取得できます。\n\n## AIにこのMCPサーバーを使用するように指示する\n\n個人的には、パフォーマンス向上のため、エージェントがリクエストごとにこのツールを呼び出すように設定することを推奨しています。エージェントのルールに以下を追加しています。\n\n```xml\n\n\nYou should call the mcp chain_of_thought tool every time you talk to the user, which generates a chain-of-thought stream which you will use to complete the user's request.\n\n\nBefore taking any action or responding to the user use the chain of thought tool as a scratchpad to:\n- List the specific rules that apply to the current request\n- Check if all required information is collected\n- Verify that the planned action complies with all policies\n- Iterate over tool results for correctness \n\nHere are some examples of what to iterate over inside the think tool:\n\nUser wants to cancel flight ABC123\n- Need to verify: user ID, reservation ID, reason\n- Check cancellation rules:\n * Is it within 24h of booking?\n * If not, check ticket class and insurance\n- Verify no segments flown or are in the past\n- Plan: collect missing info, verify rules, get confirmation\n\n\n\nUser wants to book 3 tickets to NYC with 2 checked bags each\n- Need user ID to check:\n * Membership tier for baggage allowance\n * Which payments methods exist in profile\n- Baggage calculation:\n * Economy class × 3 passengers\n * If regular member: 1 free bag each → 3 extra bags = $150\n * If silver member: 2 free bags each → 0 extra bags = $0\n * If gold member: 3 free bags each → 0 extra bags = $0\n- Payment rules to verify:\n * Max 1 travel certificate, 1 credit card, 3 gift cards\n * All payment methods must be in profile\n * Travel certificate remainder goes to waste\n- Plan:\n1. Get user ID\n2. Verify membership level for bag fees\n3. Check which payment methods in profile and if their combination is allowed\n4. Calculate total: ticket price + any bag fees\n5. Get explicit confirmation for booking\n\n\n\n```","Japanese","ja-JP",{"_27":53,"_31":57,"_36":58},{"_29":54,"_31":55,"_33":56},"Groq의 API를 사용하여 Qwen의 qwq 모델에서 원시 사고 체인 토큰을 노출하는 MCP 서버로, LLM이 응답하기 전에 단계별로 생각할 수 있도록 합니다.","Chain of Thought MCP 서버","# Chain of Thought MCP 서버\n\nAnthropic의 최근 기사 [\"생각 도구: 복잡한 도구 사용 상황에서 Claude가 멈추고 생각할 수 있도록 지원\"에서는](https://www.anthropic.com/engineering/claude-think-tool) 외부 `think` 도구를 사용하면 [SWE Bench](https://www.swebench.com/) 에서 성능이 눈에 띄게 향상된다는 것을 보여줍니다.\n\n이 MCP 서버는 Groq의 API를 사용하여 Qwen의 [qwq 모델](https://qwenlm.github.io/blog/qwq-32b/) 에서 원시 사고 체인 토큰을 노출하는 LLM을 호출합니다.\n\n## 설치\n\n1. 이 저장소를 로컬 컴퓨터에 복제합니다.\n2. 종속성을 설치하려면 `uv sync` 실행하세요.\n3. [여기](https://console.groq.com/keys) 에서 Groq API 키를 받으세요.\n4. 다음을 사용하여 mcp 구성을 업데이트하세요.\n\n지엑스피1\n\n경로는 이 저장소의 로컬 경로여야 합니다. 저장소 루트에서 터미널의 `pwd` 명령어를 실행하면 쉽게 경로를 확인할 수 있습니다.\n\n## AI에게 이 MCP 서버를 사용하도록 지시\n\n저는 개인적으로 에이전트가 모든 요청에 대해 이 도구를 호출하여 성능을 높이는 것을 선호합니다. 저는 에이전트 규칙에 다음을 추가합니다.\n\n```xml\n\n\nYou should call the mcp chain_of_thought tool every time you talk to the user, which generates a chain-of-thought stream which you will use to complete the user's request.\n\n\nBefore taking any action or responding to the user use the chain of thought tool as a scratchpad to:\n- List the specific rules that apply to the current request\n- Check if all required information is collected\n- Verify that the planned action complies with all policies\n- Iterate over tool results for correctness \n\nHere are some examples of what to iterate over inside the think tool:\n\nUser wants to cancel flight ABC123\n- Need to verify: user ID, reservation ID, reason\n- Check cancellation rules:\n * Is it within 24h of booking?\n * If not, check ticket class and insurance\n- Verify no segments flown or are in the past\n- Plan: collect missing info, verify rules, get confirmation\n\n\n\nUser wants to book 3 tickets to NYC with 2 checked bags each\n- Need user ID to check:\n * Membership tier for baggage allowance\n * Which payments methods exist in profile\n- Baggage calculation:\n * Economy class × 3 passengers\n * If regular member: 1 free bag each → 3 extra bags = $150\n * If silver member: 2 free bags each → 0 extra bags = $0\n * If gold member: 3 free bags each → 0 extra bags = $0\n- Payment rules to verify:\n * Max 1 travel certificate, 1 credit card, 3 gift cards\n * All payment methods must be in profile\n * Travel certificate remainder goes to waste\n- Plan:\n1. Get user ID\n2. Verify membership level for bag fees\n3. Check which payment methods in profile and if their combination is allowed\n4. Calculate total: ticket price + any bag fees\n5. Get explicit confirmation for booking\n\n\n\n```","Korean","ko-KR",{"_27":60,"_31":64,"_36":65},{"_29":61,"_31":62,"_33":63},"MCP 服务器使用 Groq 的 API 来公开来自 Qwen 的 qwq 模型的原始思路链标记，使 LLM 能够在响应之前逐步思考。","思路链 MCP 服务器","# 思路链 MCP 服务器\n\nAnthropic 最近发表的文章[“思考工具：让 Claude 在复杂的工具使用情况下停下来思考”](https://www.anthropic.com/engineering/claude-think-tool)表明，使用外部`think`工具可以显著提高[SWE Bench](https://www.swebench.com/)的性能。\n\n该 MCP 服务器使用 Groq 的 API 来调用 LLM，从而公开来自 Qwen 的[qwq 模型](https://qwenlm.github.io/blog/qwq-32b/)的原始思路链标记。\n\n## 安装\n\n1. 将此存储库克隆到您的本地机器。\n2. 运行`uv sync`来安装依赖项\n3. 从[这里](https://console.groq.com/keys)获取 Groq API 密钥。\n4. 使用以下命令更新您的 mcp 配置：\n\n```json\n\"mcpServers\": {\n \"chain_of_thought\": {\n \"command\": \"uv\",\n \"args\": [\n \"--directory\",\n \"path/to/cot-mcp-server\",\n \"run\",\n \"src/server.py\"\n ],\n \"env\": {\n \"GROQ_API_KEY\": \"your-groq-api-key\"\n }\n }\n}\n```\n\n该路径应为该仓库的本地路径。您可以通过在仓库根目录下的终端中运行`pwd`轻松获取该路径。\n\n## 指示 AI 使用此 MCP 服务器\n\n我个人更喜欢代理在每次请求时都调用此工具来提高性能。我将以下内容添加到代理规则中：\n\n```xml\n\n\nYou should call the mcp chain_of_thought tool every time you talk to the user, which generates a chain-of-thought stream which you will use to complete the user's request.\n\n\nBefore taking any action or responding to the user use the chain of thought tool as a scratchpad to:\n- List the specific rules that apply to the current request\n- Check if all required information is collected\n- Verify that the planned action complies with all policies\n- Iterate over tool results for correctness \n\nHere are some examples of what to iterate over inside the think tool:\n\nUser wants to cancel flight ABC123\n- Need to verify: user ID, reservation ID, reason\n- Check cancellation rules:\n * Is it within 24h of booking?\n * If not, check ticket class and insurance\n- Verify no segments flown or are in the past\n- Plan: collect missing info, verify rules, get confirmation\n\n\n\nUser wants to book 3 tickets to NYC with 2 checked bags each\n- Need user ID to check:\n * Membership tier for baggage allowance\n * Which payments methods exist in profile\n- Baggage calculation:\n * Economy class × 3 passengers\n * If regular member: 1 free bag each → 3 extra bags = $150\n * If silver member: 2 free bags each → 0 extra bags = $0\n * If gold member: 3 free bags each → 0 extra bags = $0\n- Payment rules to verify:\n * Max 1 travel certificate, 1 credit card, 3 gift cards\n * All payment methods must be in profile\n * Travel certificate remainder goes to waste\n- Plan:\n1. Get user ID\n2. Verify membership level for bag fees\n3. Check which payment methods in profile and if their combination is allowed\n4. Calculate total: ticket price + any bag fees\n5. Get explicit confirmation for booking\n\n\n\n```","Chinese","zh-CN","githubUser","locale",{"_69":70,"_13":71,"_79":30,"_80":30,"_16":81,"_82":83,"_31":32,"_84":85,"_93":-5,"_94":95,"_136":137,"_141":-5,"_92":122,"_142":127,"_124":143,"_144":145},"addedAt","2025-03-23T19:19:32.970297Z",[72,73,74,75,76,77,78],"category:autonomous-agents","category:code-execution","category:rag-systems","environment:network","hosting:remote-capable","language:python","license_grade:f","descriptionMarkdown","descriptionPlainText",3368,"integrations",[],"namespace",{"_86":87,"_92":91},"owner",{"_88":89},"githubAccount",{"_90":91},"login","beverm2391","slug","redditPostExternalId","repository",{"_96":97,"_108":109,"_112":113,"_131":-5,"_132":133,"_134":-5,"_135":-5},"argumentsJsonSchema",{"_98":99,"_105":106,"_103":107},"properties",{"_100":101},"GROQ_API_KEY",{"_29":102,"_103":104},"Your Groq API key obtained from https://console.groq.com/keys","type","string","required",[100],"object","githubProject",{"_33":34,"_110":111},"url","https://github.com/beverm2391/chain-of-thought-mcp-server","githubRepository",{"_114":115,"_116":117,"_118":119,"_120":121,"_31":122,"_86":123,"_126":127,"_128":-5,"_129":130},"defaultBranch","main","fullName","beverm2391/chain-of-thought-mcp-server","githubTreeId",102517,"language","Python","chain-of-thought-mcp-server",{"_90":91,"_124":125},"uid","5RGlmADK8t","recentStargazers",0,"spdxLicense","stargazers",11,"glamaJson","inspectable",false,"npmPackage","supportedPlatforms","scores",{"_138":127,"_139":-5,"_140":-5},"license","quality","security","simpleIcon","toolCount","dmu3x0yyz5","updatedAt","2025-09-10T11:45:01.489586Z","meta",[148,151,153,160,163,165,168,171,174,177,180,182,185,188,190,192,194,196,198],{"_149":150},"title","Chain of Thought MCP Server | Glama",{"_152":30,"_31":29},"content",{"_154":155,"_156":157,"_158":159},"href","https://glama.ai/mcp/servers/@beverm2391/chain-of-thought-mcp-server","rel","canonical","tagName","link",{"_152":32,"_161":162},"property","og:title",{"_152":30,"_161":164},"og:description",{"_152":166,"_161":167},"https://glama.ai/logo.png","og:logo",{"_152":169,"_161":170},"https://glama.ai/mcp/servers/@beverm2391/chain-of-thought-mcp-server/og-image","og:image",{"_152":172,"_161":173},"630","og:image:height",{"_152":175,"_161":176},"1200","og:image:width",{"_152":178,"_161":179},"website","og:type",{"_152":155,"_161":181},"og:url",{"_152":183,"_161":184},"Glama – MCP Hosting Platform","og:site_name",{"_152":186,"_31":187},"summary_large_image","twitter:card",{"_152":183,"_31":189},"twitter:title",{"_152":30,"_31":191},"twitter:description",{"_152":169,"_31":193},"twitter:image",{"_152":172,"_31":195},"twitter:image:height",{"_152":175,"_31":197},"twitter:image:width",{"_199":200},"script:ld+json",{"_201":202,"_203":204},"@context","http://schema.org","@graph",[205],{"_201":206,"_207":208,"_209":210},"https://schema.org","@type","BreadcrumbList","itemListElement",[211,218,222,226],{"_207":212,"_213":214,"_31":215,"_216":217},"ListItem","item","/","Glama","position",1,{"_207":212,"_213":219,"_31":220,"_216":221},"/mcp","MCP",2,{"_207":212,"_213":223,"_31":224,"_216":225},"/mcp/servers","Servers",3,{"_207":212,"_213":227,"_31":32,"_216":228},"/mcp/servers/@beverm2391/chain-of-thought-mcp-server",4,"visitorIsMaintainer","routes/_public/mcp/servers/~namespace/~slug/_pages/reviews/_route",{"_66":-5,"_27":232,"_146":247,"_283":-5},{"_69":70,"_13":233,"_79":30,"_80":30,"_16":81,"_82":234,"_31":32,"_84":235,"_93":-5,"_94":238,"_136":246,"_141":-5,"_92":122,"_142":127,"_124":143,"_144":145},[72,73,74,75,76,77,78],[],{"_86":236,"_92":91},{"_88":237},{"_90":91},{"_96":239,"_108":243,"_112":244,"_131":-5,"_132":133,"_134":-5,"_135":-5},{"_98":240,"_105":242,"_103":107},{"_100":241},{"_29":102,"_103":104},[100],{"_33":34,"_110":111},{"_114":115,"_116":117,"_118":119,"_120":121,"_31":122,"_86":245,"_126":127,"_128":-5,"_129":130},{"_90":91,"_124":125},{"_138":127,"_139":-5,"_140":-5},[248,250,252,254,256,257,258,259,260,261,262,263,264,265,266,267,268,269,270],{"_149":249},"Reviews | Chain of Thought MCP Server | Glama",{"_152":251,"_31":29},"Read community reviews and feedback for Chain of Thought MCP Server. Discover user experiences, ratings, strengths, and potential limitations of this MCP implementation from actual users.",{"_154":253,"_156":157,"_158":159},"https://glama.ai/mcp/servers/@beverm2391/chain-of-thought-mcp-server/reviews",{"_152":255,"_161":162},"Reviews | Chain of Thought MCP Server",{"_152":251,"_161":164},{"_152":166,"_161":167},{"_152":169,"_161":170},{"_152":172,"_161":173},{"_152":175,"_161":176},{"_152":178,"_161":179},{"_152":253,"_161":181},{"_152":183,"_161":184},{"_152":186,"_31":187},{"_152":183,"_31":189},{"_152":251,"_31":191},{"_152":169,"_31":193},{"_152":172,"_31":195},{"_152":175,"_31":197},{"_199":271},{"_201":202,"_203":272},[273],{"_201":206,"_207":208,"_209":274},[275,276,277,278,279],{"_207":212,"_213":214,"_31":215,"_216":217},{"_207":212,"_213":219,"_31":220,"_216":221},{"_207":212,"_213":223,"_31":224,"_216":225},{"_207":212,"_213":227,"_31":32,"_216":228},{"_207":212,"_213":280,"_31":281,"_216":282},"/mcp/servers/@beverm2391/chain-of-thought-mcp-server/reviews","Reviews",5,"review","actionData","errors"]

MCP directory API