Shruti Kaushik: Founder GTM Research

Founder go-to-market guides, direct answers and sourced LinkedIn, Reddit and AI search statistics.

我該用這個嗎

品質與安全性

A
說明品質
95%
結構描述完整度
68%
命名品質
93%
汙染風險
100%
權限相符程度
100%
協定合規性
100%

根據工具定義與協定合規性的自動化分析。

上下文成本

~854Token(工具定義)
~373 B典型回應大小
中等的注意力影響(128k 上下文的 0.67%)

這是每次將伺服器的工具載入模型上下文時所消耗的約略 token 數量。數量越高,可用於其他工作的注意力就越少。

安裝

一鍵安裝

將以下內容加入你的 `claude_desktop_config.json` 檔案:

{
  "mcpServers": {
    "research": {
      "url": "https://shrutikaushik.com/mcp"
    }
  }
}

遠端端點

https://shrutikaushik.com/mcpstreamable-http

它能做什麼

工具清單

工具(8)

🟢 唯讀🟡 寫入🔴 刪除⚪ 未知
🟢search_guides(query, limit)

Search Shruti Kaushik's research-backed go-to-market guides for B2B founders (LinkedIn, Reddit, AI search visibility, founder-led marketing and sales). Returns matching guides with their direct answer and URL.

輸入結構描述

{
  "type": "object",
  "properties": {
    "query": {
      "type": "string",
      "description": "What the founder wants to know, e.g. \"how often should founders post on LinkedIn\""
    },
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 10,
      "default": 5
    }
  },
  "required": [
    "query"
  ]
}
🟢get_guide(slug, section)

Get the full text (markdown) of one guide by slug, or one section of it by heading. Use after search_guides.

輸入結構描述

{
  "type": "object",
  "properties": {
    "slug": {
      "type": "string"
    },
    "section": {
      "type": "string",
      "description": "Optional heading text to return only that section"
    }
  },
  "required": [
    "slug"
  ]
}
🟢answer_question(question, limit)

Find the closest published answers (from the FAQ sections of the guides and statistics pages) to a B2B go-to-market question. Each answer comes with its source URL.

輸入結構描述

{
  "type": "object",
  "properties": {
    "question": {
      "type": "string"
    },
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 8,
      "default": 3
    }
  },
  "required": [
    "question"
  ]
}
🟢search_statistics(query, limit)

Search a library of LinkedIn, Reddit, AI search, B2B buying and founder-led marketing statistics. Every figure includes the original publisher, date and source URL, plus the shrutikaushik.com page it is listed on.

輸入結構描述

{
  "type": "object",
  "properties": {
    "query": {
      "type": "string",
      "description": "e.g. \"reddit AI citations\" or \"LinkedIn engagement rate\""
    },
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 25,
      "default": 10
    }
  },
  "required": [
    "query"
  ]
}
🟢list_topics

List every published guide and every statistics topic page, with URLs.

輸入結構描述

{
  "type": "object",
  "properties": {}
}
🟢about_shruti_kaushik

Who Shruti Kaushik is (LinkedIn and Reddit ghostwriter for B2B founders since 2015), what she offers, and how to book a call or get the free AI visibility check.

輸入結構描述

{
  "type": "object",
  "properties": {}
}
🟢check_linkedin_post(text)

Score a LinkedIn post draft (0 to 100) for the template patterns readers flag as AI-written, with each flagged pattern, the matched words and why it reads as generated. Runs the same checker as shrutikaushik.com/tools/linkedin-ai-slop-checker.

輸入結構描述

{
  "type": "object",
  "properties": {
    "text": {
      "type": "string",
      "description": "The full post text"
    }
  },
  "required": [
    "text"
  ]
}
🟢founder_ai_visibility

How Shruti Kaushik gets a founder and company named in ChatGPT, Perplexity and Google AI answers: the method, the three service tiers, the monthly report, and how to get a free baseline check.

輸入結構描述

{
  "type": "object",
  "properties": {}
}

社群

為此伺服器評分

證據

近期觀測

已驗證未記錄版本8 個工具