Kindly Web Search MCP Server

Kindly Web Search is an MCP server designed to provide AI coding assistants with up-to-date API documentation and robust web content retrieval. Unlike generic search MCP servers that only return snippets, Kindly retrieves full conversations (questions, answers, comments, and metadata) from specialized sources like StackOverflow, GitHub, and arXiv, or parses arbitrary webpages using a headless browser.

It uses a priority-based content resolver to fetch data efficiently:

1. StackExchange API (StackOverflow, etc.)
2. GitHub GraphQL API (Issues and Discussions)
3. MediaWiki Action API (Wikipedia)
4. Atom API + PDF to Markdown (arXiv)
5. Universal HTML loader (Headless Chromium via nodriver for all other URLs)

It also features 'Markdown fast paths' to skip the browser for documentation sites that support direct Markdown fetching via specific host suffixes or Accept: text/markdown headers.

To use the Kindly Web Search MCP server with Cursor, create a .cursor/mcp.json file in your project.

Note on Timeouts

Cursor does not currently support per-server startup timeouts. If the initial run is slow, run the uvx command from the Quickstart section once in a terminal to prebuild the environment, then restart Cursor.

{
  "mcpServers": {
    "kindly-web-search": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "--from",
        "git+https://github.com/Shelpuk-AI-Technology-Consulting/kindly-web-search-mcp-server",
        "kindly-web-search-mcp-server",
        "start-mcp-server"
      ],
      "env": {
        "SERPER_API_KEY": "${env:SERPER_API_KEY}",
        "TAVILY_API_KEY": "${env:TAVILY_API_KEY}",
        "SEARXNG_BASE_URL": "${env:SEARXNG_BASE_URL}",
        "GITHUB_TOKEN": "${env:GITHUB_TOKEN}",
        "KINDLY_BROWSER_EXECUTABLE_PATH": "${env:KINDLY_BROWSER_EXECUTABLE_PATH}"
      }
    }
  }
}

To use the Kindly Web Search MCP server with OpenClaw, it is recommended to use mcporter via the CLI.

CLI Installation

Use mcporter config add to add the server. This writes to ~/.mcporter/mcporter.json when using --scope home.

File-based Configuration

Alternatively, edit your mcporter.json file directly.

Warning: Do not add mcpServers to the root level of ~/.openclaw/openclaw.json, as OpenClaw uses strict schema validation and will reject unknown keys. If the server is not detected, run openclaw gateway restart.

Verification

Verify the configuration using mcporter config get <server-name>.

# Install mcporter if not present
npm i -g mcporter

# Add server via CLI
mcporter config add kindly-search \
  --scope home \
  --command "uvx --from git+https://github.com/Shelpuk-AI-Technology-Consulting/kindly-web-search-mcp-server kindly-web-search-mcp-server start-mcp-server" \
  --env SERPER_API_KEY="$SERPER_API_KEY" \
  --env TAVILY_API_KEY="$TAVILY_API_KEY" \
  --env SEARXNG_BASE_URL="$SEARXNG_BASE_URL" \
  --env GITHUB_TOKEN="$GITHUB_TOKEN" \
  --env KINDLY_BROWSER_EXECUTABLE_PATH="$KINDLY_BROWSER_EXECUTABLE_PATH"

To use the Kindly Web Search MCP server with Claude Desktop, edit your claude_desktop_config.json file.

File Locations:

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json

Important Notes

• Values in this file are literal strings. Do not commit or share this file.
• Claude Desktop does not support per-server startup timeouts. If the first run is slow, run the uvx command from the Quickstart section once in a terminal to prebuild the environment, then restart Claude Desktop.

{
  "mcpServers": {
    "kindly-web-search": {
      "command": "uvx",
      "args": [
        "--from",
        "git+https://github.com/Shelpuk-AI-Technology-Consulting/kindly-web-search-mcp-server",
        "kindly-web-search-mcp-server",
        "start-mcp-server"
      ],
      "env": {
        "SERPER_API_KEY": "PASTE_SERPER_KEY_OR_LEAVE_EMPTY",
        "TAVILY_API_KEY": "PASTE_TAVILY_KEY_OR_LEAVE_EMPTY",
        "SEARXNG_BASE_URL": "PASTE_SEARXNG_URL_OR_LEAVE_EMPTY",
        "GITHUB_TOKEN": "PASTE_GITHUB_TOKEN_OR_LEAVE_EMPTY",
        "KINDLY_BROWSER_EXECUTABLE_PATH": "PASTE_IF_NEEDED"
      }
    }
  }
}

To add the Kindly Web Search MCP server to Antigravity:

1. Open the MCP store in Antigravity.
2. Click Manage MCP Servers.
3. Click View raw config to open mcp_config.json.
4. Add the server configuration under the mcpServers object.
5. Save the file and click Refresh.

Troubleshooting

• If uvx is not found, use its absolute path (e.g., via which uvx on macOS/Linux or where uvx on Windows).
• If the first startup is slow, run the uvx command from the Quickstart section once in a terminal to prebuild the environment, then click Refresh.
• Security Note: Do not commit or share mcp_config.json if it contains sensitive API keys.

{
  "kindly-web-search": {
    "command": "uvx",
    "args": [
      "--from",
      "git+https://github.com/Shelpuk-AI-Technology-Consulting/kindly-web-search-mcp-server",
      "kindly-web-search-mcp-server",
      "start-mcp-server"
    ],
    "env": {
      "SERPER_API_KEY": "PASTE_SERPER_KEY_OR_LEAVE_EMPTY",
      "TAVILY_API_KEY": "PASTE_TAVILY_KEY_OR_LEAVE_EMPTY",
      "SEARXNG_BASE_URL": "PASTE_SEARXNG_URL_OR_LEAVE_EMPTY",
      "GITHUB_TOKEN": "PASTE_GITHUB_TOKEN_OR_LEAVE_EMPTY",
      "KINDLY_BROWSER_EXECUTABLE_PATH": "PASTE_IF_NEEDED"
    }
  }
}

To use the Kindly Web Search MCP server with Gemini CLI, edit your ~/.gemini/settings.json (or .gemini/settings.json in your project directory). You must provide one of the search provider credentials: SERPER_API_KEY, SERPBASE_API_KEY, TAVILY_API_KEY, SEARXNG_BASE_URL, or SOFYA_API_KEY.

{
  "mcpServers": {
    "kindly-web-search": {
      "command": "uvx",
      "args": [
        "--from",
        "git+https://github.com/Shelpuk-AI-Technology-Consulting/kindly-web-search-mcp-server",
        "kindly-web-search-mcp-server",
        "start-mcp-server"
      ],
      "env": {
        "SERPER_API_KEY": "$SERPER_API_KEY",
        "TAVILY_API_KEY": "$TAVILY_API_KEY",
        "SEARXNG_BASE_URL": "$SEARXNG_BASE_URL",
        "GITHUB_TOKEN": "$GITHUB_TOKEN",
        "KINDLY_BROWSER_EXECUTABLE_PATH": "$KINDLY_BROWSER_EXECUTABLE_PATH"
      },
      "timeout": 120000
    }
  }
}

The most secure way to configure the Kindly Web Search MCP server in VS Code (GitHub Copilot / Microsoft Copilot) is to use interactive prompts. This avoids storing secrets in plain text files.

Create a .vscode/mcp.json file in your project and define the server and its required inputs. This will trigger prompts for the API keys and other configuration values when the server starts.

{
  "servers": {
    "kindly-web-search": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "--from",
        "git+https://github.com/Shelpuk-AI-Technology-Consulting/kindly-web-search-mcp-server",
        "kindly-web-search-mcp-server",
        "start-mcp-server"
      ],
      "env": {
        "SERPER_API_KEY": "${input:serper-api-key}",
        "TAVILY_API_KEY": "${input:tavily-api-key}",
        "SEARXNG_BASE_URL": "${input:searxng-base-url}",
        "GITHUB_TOKEN": "${input:github-token}",
        "KINDLY_BROWSER_EXECUTABLE_PATH": "${input:browser-path}"
      }
    }
  },
  "inputs": [
    { "id": "serper-api-key", "type": "promptString", "description": "Serper API key (optional if using Tavily or SearXNG)" },
    { "id": "tavily-api-key", "type": "promptString", "description": "Tavily API key (optional if using Serper or SearXNG)" },
    { "id": "searxng-base-url", "type": "promptString", "description": "SearXNG base URL (optional if using Serper or Tavily)" },
    { "id": "github-token", "type": "promptString", "description": "GitHub token (recommended)" },
    { "id": "browser-path", "type": "promptString", "description": "Browser binary path (only if needed)" }
  ]
}

To use the Kindly Web Search MCP server with Claude Code, provide one of the search provider credentials: SERPER_API_KEY, SERPBASE_API_KEY, TAVILY_API_KEY, SEARXNG_BASE_URL, or SOFYA_API_KEY.

CLI Installation

Use the claude mcp add command. Note that for current Claude Code versions, the server name must be placed immediately after --transport stdio and before any -e or --env flags.

File-based Configuration

Create or edit a .mcp.json file (recommended for project-specific scope and teams).

Handling Timeouts

If Claude Code times out during startup, increase the timeout by setting the MCP_TIMEOUT environment variable (value in milliseconds).

# macOS / Linux (Serper example)
claude mcp add --transport stdio kindly-web-search \
  -e SERPER_API_KEY="$SERPER_API_KEY" \
  -e GITHUB_TOKEN="$GITHUB_TOKEN" \
  -e KINDLY_BROWSER_EXECUTABLE_PATH="$KINDLY_BROWSER_EXECUTABLE_PATH" \
  -- uvx --from git+https://github.com/Shelpuk-AI-Technology-Consulting/kindly-web-search-mcp-server \
  kindly-web-search-mcp-server start-mcp-server

# Increase timeout if needed
export MCP_TIMEOUT=120000