Markdown MCP Server

get_page_markdown

Extract clean markdown from web pages by removing navigation, headers, and sidebars while preserving main content and formatting for readability.

Instructions

Extract clean markdown content from a URL. Returns only the main content without navigation, headers, footers, or sidebars.

Input Schema

TableJSON Schema

Name	Required	Description
`url`	Yes	The URL to extract markdown from
`includeImages`	No	Whether to include image references in markdown (default: true)
`includeLinks`	No	Whether to include hyperlinks in markdown (default: true)
`waitForSelector`	No	Optional CSS selector to wait for before extracting content
`timeout`	No	Navigation timeout in milliseconds (default: 30000)

Implementation Reference

markdown-mcp.js:86-432 (handler)

Main implementation of the get_page_markdown tool handler. Uses Playwright to load the page, evaluate JavaScript to find main content, skip navigation/UI elements, and convert HTML to clean Markdown format.

async getPageMarkdown(args) {
  const {
    url,
    includeImages = true,
    includeLinks = true,
    waitForSelector,
    timeout = 30000,
  } = args;

  const browser = await this.ensureBrowser();
  const context = await browser.newContext();
  const page = await context.newPage();

  try {
    await page.goto(url, { waitUntil: 'domcontentloaded', timeout });

    if (waitForSelector) {
      await page.waitForSelector(waitForSelector, { timeout: 10000 });
    } else {
      // Wait for content to load - especially important for JS-heavy sites
      await page.waitForTimeout(5000);
    }

    const markdown = await page.evaluate(
      ({ includeImages, includeLinks }) => {
        function extractMainContent() {
          // Confluence-specific selectors first, then general ones
          const mainSelectors = [
            '#main-content',
            '.wiki-content',
            '[data-test-id="wiki-content"]',
            'main[role="main"]',
            'main',
            'article',
            '[role="main"]',
            '.main-content',
            '.content',
            '#content',
            '.post-content',
            '.article-content',
            'body',
          ];

          for (const selector of mainSelectors) {
            const element = document.querySelector(selector);
            if (element && element.textContent.trim().length > 100) {
              return element;
            }
          }

          return document.body;
        }

        function shouldSkipElement(element) {
          if (!element || !element.tagName) return true;

          const tagName = element.tagName.toLowerCase();
          
          // Never skip these content elements
          if (['p', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'ul', 'ol', 'li', 'table', 'pre', 'code', 'blockquote'].includes(tagName)) {
            return false;
          }

          // Check for hidden elements
          if (element.offsetParent === null && tagName !== 'script' && tagName !== 'style') {
            const style = window.getComputedStyle(element);
            if (style.display === 'none' || style.visibility === 'hidden') {
              return true;
            }
          }

          // Skip technical elements
          if (['script', 'style', 'noscript', 'iframe'].includes(tagName)) {
            return true;
          }

          // Check role attributes
          const role = element.getAttribute('role');
          if (['navigation', 'banner', 'contentinfo', 'complementary'].includes(role)) {
            return true;
          }

          // Check specific element types
          if (tagName === 'nav' || tagName === 'header' || tagName === 'footer' || tagName === 'aside') {
            return true;
          }

          // Check class and id for common patterns (but be less aggressive)
          const className = (element.className || '').toString().toLowerCase();
          const id = (element.id || '').toLowerCase();
          const combined = className + ' ' + id;

          const strictSkipPatterns = [
            'cookie-banner',
            'gdpr',
            'advertisement',
            'sponsored',
          ];

          return strictSkipPatterns.some(pattern => combined.includes(pattern));
        }

        function getTextContent(node) {
          let text = '';
          for (const child of node.childNodes) {
            if (child.nodeType === Node.TEXT_NODE) {
              text += child.textContent;
            } else if (child.nodeType === Node.ELEMENT_NODE) {
              const tag = child.tagName.toLowerCase();
              if (tag === 'br') {
                text += '\n';
              } else if (!shouldSkipElement(child)) {
                text += getTextContent(child);
              }
            }
          }
          return text;
        }

        function convertToMarkdown(node, depth = 0, inList = false) {
          if (!node || shouldSkipElement(node)) return '';

          let markdown = '';
          const tagName = node.tagName?.toLowerCase();

          // Headings
          if (tagName?.match(/^h[1-6]$/)) {
            const level = parseInt(tagName[1]);
            const text = getTextContent(node).trim();
            if (text) {
              markdown += '\n' + '#'.repeat(level) + ' ' + text + '\n\n';
            }
            return markdown;
          }

          // Paragraphs
          if (tagName === 'p') {
            let content = '';
            for (const child of node.childNodes) {
              if (child.nodeType === Node.TEXT_NODE) {
                content += child.textContent;
              } else if (child.nodeType === Node.ELEMENT_NODE) {
                content += convertToMarkdown(child, depth + 1);
              }
            }
            const text = content.trim();
            if (text) {
              markdown += text + '\n\n';
            }
            return markdown;
          }

          // Code blocks
          if (tagName === 'pre') {
            const code = node.querySelector('code');
            const text = (code || node).textContent.trim();
            if (text) {
              const language = code?.className.match(/language-(\w+)/)?.[1] || '';
              markdown += '\n```' + language + '\n' + text + '\n```\n\n';
            }
            return markdown;
          }

          // Inline code
          if (tagName === 'code' && node.parentElement?.tagName !== 'PRE') {
            return '`' + node.textContent.trim() + '`';
          }

          // Blockquotes
          if (tagName === 'blockquote') {
            const text = getTextContent(node).trim();
            if (text) {
              const lines = text.split('\n').filter(l => l.trim());
              markdown += '\n' + lines.map(line => '> ' + line.trim()).join('\n') + '\n\n';
            }
            return markdown;
          }

          // Lists
          if (tagName === 'ul' || tagName === 'ol') {
            const items = Array.from(node.children).filter(child => child.tagName === 'LI');
            items.forEach((li, idx) => {
              const prefix = tagName === 'ol' ? `${idx + 1}. ` : '- ';
              let itemContent = '';
              for (const child of li.childNodes) {
                if (child.nodeType === Node.TEXT_NODE) {
                  itemContent += child.textContent;
                } else if (child.nodeType === Node.ELEMENT_NODE) {
                  itemContent += convertToMarkdown(child, depth + 1, true);
                }
              }
              const text = itemContent.trim();
              if (text) {
                markdown += prefix + text + '\n';
              }
            });
            if (!inList) markdown += '\n';
            return markdown;
          }

          // Images
          if (tagName === 'img' && includeImages) {
            const alt = node.getAttribute('alt') || '';
            const src = node.getAttribute('src') || node.getAttribute('data-src') || '';
            if (src) {
              try {
                const fullSrc = new URL(src, window.location.href).href;
                markdown += `![${alt}](${fullSrc})\n\n`;
              } catch (e) {
                // Invalid URL, skip
              }
            }
            return markdown;
          }

          // Links
          if (tagName === 'a' && includeLinks) {
            const text = getTextContent(node).trim();
            const href = node.getAttribute('href');
            if (text && href) {
              try {
                const fullHref = new URL(href, window.location.href).href;
                return `[${text}](${fullHref})`;
              } catch (e) {
                return text;
              }
            }
            return text || '';
          }

          // Strong/Bold
          if (tagName === 'strong' || tagName === 'b') {
            const text = getTextContent(node).trim();
            return text ? `**${text}**` : '';
          }

          // Emphasis/Italic
          if (tagName === 'em' || tagName === 'i') {
            const text = getTextContent(node).trim();
            return text ? `*${text}*` : '';
          }

          // Horizontal rule
          if (tagName === 'hr') {
            return '\n---\n\n';
          }

          // Tables
          if (tagName === 'table') {
            const rows = Array.from(node.querySelectorAll('tr'));
            if (rows.length > 0) {
              rows.forEach((row, rowIdx) => {
                const cells = Array.from(row.querySelectorAll('th, td'));
                const cellTexts = cells.map(cell => getTextContent(cell).trim().replace(/\n/g, ' '));
                if (cellTexts.some(t => t)) {
                  markdown += '| ' + cellTexts.join(' | ') + ' |\n';
                  if (rowIdx === 0) {
                    markdown += '| ' + cells.map(() => '---').join(' | ') + ' |\n';
                  }
                }
              });
              markdown += '\n';
            }
            return markdown;
          }

          // Line break
          if (tagName === 'br') {
            return '\n';
          }

          // Container elements - process children
          if (['div', 'section', 'article', 'main', 'span', 'td', 'th', 'li'].includes(tagName)) {
            for (const child of node.childNodes) {
              if (child.nodeType === Node.ELEMENT_NODE) {
                markdown += convertToMarkdown(child, depth + 1, inList);
              } else if (child.nodeType === Node.TEXT_NODE && depth === 0 && !inList) {
                const text = child.textContent.trim();
                if (text && text.length > 0) {
                  markdown += text + ' ';
                }
              }
            }
            return markdown;
          }

          // For any other element, try to extract text from children
          if (node.childNodes && node.childNodes.length > 0) {
            for (const child of node.childNodes) {
              if (child.nodeType === Node.ELEMENT_NODE) {
                markdown += convertToMarkdown(child, depth + 1, inList);
              }
            }
          }

          return markdown;
        }

        const mainContent = extractMainContent();
        let result = convertToMarkdown(mainContent);

        // Clean up excessive newlines and spaces
        result = result
          .replace(/ +/g, ' ')  // Multiple spaces to single
          .replace(/\n\n\n+/g, '\n\n')  // Multiple newlines to double
          .trim();

        // If still empty, use fallback
        if (!result || result.length < 50) {
          const allText = mainContent.textContent.trim();
          if (allText) {
            result = allText
              .split('\n')
              .map(line => line.trim())
              .filter(line => line.length > 0)
              .join('\n\n');
          }
        }

        return result;
      },
      { includeImages, includeLinks }
    );

    await context.close();

    return {
      content: [
        {
          type: 'text',
          text: markdown || 'No content could be extracted from this page.',
        },
      ],
    };
  } catch (error) {
    await context.close();
    return {
      content: [
        {
          type: 'text',
          text: `Error extracting markdown: ${error.message}`,
        },
      ],
      isError: true,
    };
  }
}

markdown-mcp.js:36-64 (schema)

Input schema definition for the get_page_markdown tool, specifying parameters like url (required), includeImages, includeLinks, waitForSelector, and timeout.

inputSchema: {
  type: 'object',
  properties: {
    url: {
      type: 'string',
      description: 'The URL to extract markdown from',
    },
    includeImages: {
      type: 'boolean',
      description: 'Whether to include image references in markdown (default: true)',
      default: true,
    },
    includeLinks: {
      type: 'boolean',
      description: 'Whether to include hyperlinks in markdown (default: true)',
      default: true,
    },
    waitForSelector: {
      type: 'string',
      description: 'Optional CSS selector to wait for before extracting content',
    },
    timeout: {
      type: 'number',
      description: 'Navigation timeout in milliseconds (default: 30000)',
      default: 30000,
    },
  },
  required: ['url'],
},

markdown-mcp.js:31-67 (registration)

Registration of the get_page_markdown tool in the ListToolsRequest handler, providing name, description, and input schema.

this.server.setRequestHandler(ListToolsRequestSchema, async () => ({
  tools: [
    {
      name: 'get_page_markdown',
      description: 'Extract clean markdown content from a URL. Returns only the main content without navigation, headers, footers, or sidebars.',
      inputSchema: {
        type: 'object',
        properties: {
          url: {
            type: 'string',
            description: 'The URL to extract markdown from',
          },
          includeImages: {
            type: 'boolean',
            description: 'Whether to include image references in markdown (default: true)',
            default: true,
          },
          includeLinks: {
            type: 'boolean',
            description: 'Whether to include hyperlinks in markdown (default: true)',
            default: true,
          },
          waitForSelector: {
            type: 'string',
            description: 'Optional CSS selector to wait for before extracting content',
          },
          timeout: {
            type: 'number',
            description: 'Navigation timeout in milliseconds (default: 30000)',
            default: 30000,
          },
        },
        required: ['url'],
      },
    },
  ],
}));

markdown-mcp.js:69-74 (registration)

Registration of the CallToolRequest handler that dispatches calls to 'get_page_markdown' to the getPageMarkdown method.

this.server.setRequestHandler(CallToolRequestSchema, async (request) => {
  if (request.params.name === 'get_page_markdown') {
    return await this.getPageMarkdown(request.params.arguments);
  }
  throw new Error(`Unknown tool: ${request.params.name}`);
});

markdown-mcp-gemini.js:39-341 (handler)

Alternative implementation of the core extraction logic in markdown-mcp-gemini.js, used by the tool handler.

async function extractMarkdownContent(url, options = {}) {
  const {
    includeImages = true,
    includeLinks = true,
    waitForSelector,
    timeout = 30000,
  } = options;

  const browserInstance = await ensureBrowser();
  const context = await browserInstance.newContext({
    userAgent: 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36'
  });
  const page = await context.newPage();

  try {
    await page.goto(url, { waitUntil: 'domcontentloaded', timeout });

    if (waitForSelector) {
      await page.waitForSelector(waitForSelector, { timeout: 10000 });
    } else {
      // Wait for content to load - especially important for JS-heavy sites
      await page.waitForTimeout(5000);
    }

    const markdown = await page.evaluate(
      ({ includeImages, includeLinks }) => {
        function extractMainContent() {
          // Confluence-specific selectors first, then general ones
          const mainSelectors = [
            '#main-content',
            '.wiki-content',
            '[data-test-id="wiki-content"]',
            'main[role="main"]',
            'main',
            'article',
            '[role="main"]',
            '.main-content',
            '.content',
            '#content',
            '.post-content',
            '.article-content',
            'body',
          ];

          for (const selector of mainSelectors) {
            const element = document.querySelector(selector);
            if (element && element.textContent.trim().length > 100) {
              return element;
            }
          }

          return document.body;
        }

        function shouldSkipElement(element) {
          if (!element || !element.tagName) return true;

          const tagName = element.tagName.toLowerCase();
          
          // Never skip these content elements
          if (['p', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'ul', 'ol', 'li', 'table', 'pre', 'code', 'blockquote'].includes(tagName)) {
            return false;
          }

          // Check for hidden elements
          if (element.offsetParent === null && tagName !== 'script' && tagName !== 'style') {
            const style = window.getComputedStyle(element);
            if (style.display === 'none' || style.visibility === 'hidden') {
              return true;
            }
          }

          // Skip technical elements
          if (['script', 'style', 'noscript', 'iframe'].includes(tagName)) {
            return true;
          }

          // Check role attributes
          const role = element.getAttribute('role');
          if (['navigation', 'banner', 'contentinfo', 'complementary'].includes(role)) {
            return true;
          }

          // Check specific element types
          if (tagName === 'nav' || tagName === 'header' || tagName === 'footer' || tagName === 'aside') {
            return true;
          }

          // Check class and id for common patterns (but be less aggressive)
          const className = (element.className || '').toString().toLowerCase();
          const id = (element.id || '').toLowerCase();
          const combined = className + ' ' + id;

          const strictSkipPatterns = [
            'cookie-banner',
            'gdpr',
            'advertisement',
            'sponsored',
          ];

          return strictSkipPatterns.some(pattern => combined.includes(pattern));
        }

        function getTextContent(node) {
          let text = '';
          for (const child of node.childNodes) {
            if (child.nodeType === Node.TEXT_NODE) {
              text += child.textContent;
            } else if (child.nodeType === Node.ELEMENT_NODE) {
              const tag = child.tagName.toLowerCase();
              if (tag === 'br') {
                text += '\n';
              } else if (!shouldSkipElement(child)) {
                text += getTextContent(child);
              }
            }
          }
          return text;
        }

        function convertToMarkdown(node, depth = 0, inList = false) {
          if (!node || shouldSkipElement(node)) return '';

          let markdown = '';
          const tagName = node.tagName?.toLowerCase();

          // Headings
          if (tagName?.match(/^h[1-6]$/)) {
            const level = parseInt(tagName[1]);
            const text = getTextContent(node).trim();
            if (text) {
              markdown += '\n' + '#'.repeat(level) + ' ' + text + '\n\n';
            }
            return markdown;
          }

          // Paragraphs
          if (tagName === 'p') {
            let content = '';
            for (const child of node.childNodes) {
              if (child.nodeType === Node.TEXT_NODE) {
                content += child.textContent;
              } else if (child.nodeType === Node.ELEMENT_NODE) {
                content += convertToMarkdown(child, depth + 1);
              }
            }
            const text = content.trim();
            if (text) {
              markdown += text + '\n\n';
            }
            return markdown;
          }

          // Code blocks
          if (tagName === 'pre') {
            const code = node.querySelector('code');
            const text = (code || node).textContent.trim();
            if (text) {
              const language = code?.className.match(/language-(\w+)/)?.[1] || '';
              markdown += '\n```' + language + '\n' + text + '\n```\n\n';
            }
            return markdown;
          }

          // Inline code
          if (tagName === 'code' && node.parentElement?.tagName !== 'PRE') {
            return '`' + node.textContent.trim() + '`';
          }

          // Blockquotes
          if (tagName === 'blockquote') {
            const text = getTextContent(node).trim();
            if (text) {
              const lines = text.split('\n').filter(l => l.trim());
              markdown += '\n' + lines.map(line => '> ' + line.trim()).join('\n') + '\n\n';
            }
            return markdown;
          }

          // Lists
          if (tagName === 'ul' || tagName === 'ol') {
            const items = Array.from(node.children).filter(child => child.tagName === 'LI');
            items.forEach((li, idx) => {
              const prefix = tagName === 'ol' ? `${idx + 1}. ` : '- ';
              let itemContent = '';
              for (const child of li.childNodes) {
                if (child.nodeType === Node.TEXT_NODE) {
                  itemContent += child.textContent;
                } else if (child.nodeType === Node.ELEMENT_NODE) {
                  itemContent += convertToMarkdown(child, depth + 1, true);
                }
              }
              const text = itemContent.trim();
              if (text) {
                markdown += prefix + text + '\n';
              }
            });
            if (!inList) markdown += '\n';
            return markdown;
          }

          // Strong/Bold
          if (tagName === 'strong' || tagName === 'b') {
            const text = getTextContent(node).trim();
            return text ? `**${text}**` : '';
          }

          // Emphasis/Italic
          if (tagName === 'em' || tagName === 'i') {
            const text = getTextContent(node).trim();
            return text ? `*${text}*` : '';
          }

          // Horizontal rule
          if (tagName === 'hr') {
            return '\n---\n\n';
          }

          // Tables
          if (tagName === 'table') {
            const rows = Array.from(node.querySelectorAll('tr'));
            if (rows.length > 0) {
              rows.forEach((row, rowIdx) => {
                const cells = Array.from(row.querySelectorAll('th, td'));
                const cellTexts = cells.map(cell => getTextContent(cell).trim().replace(/\n/g, ' '));
                if (cellTexts.some(t => t)) {
                  markdown += '| ' + cellTexts.join(' | ') + ' |\n';
                  if (rowIdx === 0) {
                    markdown += '| ' + cells.map(() => '---').join(' | ') + ' |\n';
                  }
                }
              });
              markdown += '\n';
            }
            return markdown;
          }

          // Line break
          if (tagName === 'br') {
            return '\n';
          }

          // Container elements - process children
          if (['div', 'section', 'article', 'main', 'span', 'td', 'th', 'li'].includes(tagName)) {
            for (const child of node.childNodes) {
              if (child.nodeType === Node.ELEMENT_NODE) {
                markdown += convertToMarkdown(child, depth + 1, inList);
              } else if (child.nodeType === Node.TEXT_NODE && depth === 0 && !inList) {
                const text = child.textContent.trim();
                if (text && text.length > 0) {
                  markdown += text + ' ';
                }
              }
            }
            return markdown;
          }

          // For any other element, try to extract text from children
          if (node.childNodes && node.childNodes.length > 0) {
            for (const child of node.childNodes) {
              if (child.nodeType === Node.ELEMENT_NODE) {
                markdown += convertToMarkdown(child, depth + 1, inList);
              }
            }
          }

          return markdown;
        }

        const mainContent = extractMainContent();
        let result = convertToMarkdown(mainContent);

        // Clean up excessive newlines and spaces
        result = result
          .replace(/ +/g, ' ')  // Multiple spaces to single
          .replace(/\n\n\n+/g, '\n\n')  // Multiple newlines to double
          .trim();

        // If still empty, use fallback
        if (!result || result.length < 50) {
          const allText = mainContent.textContent.trim();
          if (allText) {
            result = allText
              .split('\n')
              .map(line => line.trim())
              .filter(line => line.length > 0)
              .join('\n\n');
          }
        }

        return result;
      },
      { includeImages, includeLinks }
    );

    await context.close();
    return markdown || 'No content could be extracted from this page.';

  } catch (error) {
    await context.close();
    throw new Error(`Error extracting markdown: ${error.message}`);
  }
}

Tool Definition Quality

A3.7/5.0

Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries the full burden of behavioral disclosure. It describes the tool's behavior by specifying what content is excluded and mentions optional parameters like waiting for selectors and timeouts, which adds useful context. However, it lacks details on error handling, rate limits, authentication needs, or output format, leaving some behavioral aspects unclear.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is highly concise and front-loaded, consisting of just two sentences that directly state the tool's purpose and key exclusions. Every sentence adds value without redundancy, making it efficient and easy to understand at a glance.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's moderate complexity (5 parameters, no output schema, no annotations), the description is somewhat complete by explaining the extraction scope and exclusions. However, it lacks details on output format, error cases, or performance considerations, which would be helpful for an AI agent to use it effectively. The absence of an output schema increases the need for more completeness in the description.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, meaning all parameters are documented in the schema itself. The description does not add any parameter-specific semantics beyond what the schema provides, such as explaining the 'waitForSelector' or 'timeout' in more detail. However, it implies the tool's focus on 'clean markdown' extraction, which contextualizes the parameters but doesn't enhance their individual meanings.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool's purpose with specific verbs ('extract clean markdown content') and resources ('from a URL'), and distinguishes what it does by specifying what it excludes ('without navigation, headers, footers, or sidebars'). It explicitly defines the scope of extraction as 'only the main content,' making the purpose unambiguous and well-differentiated.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description implies usage by stating it extracts 'clean markdown content' from URLs, suggesting it's for content extraction tasks. However, it provides no explicit guidance on when to use this tool versus alternatives, prerequisites, or exclusions. With no sibling tools mentioned, the lack of comparative guidance is less critical but still leaves usage context somewhat vague.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

get_page_markdownA

Latest Blog Posts

Lightport: Open-Sourcing Glama's AI Gateway
By punkpeye on April 27, 2026.
open source
OpenAI
Tool Definition Quality Score (TDQS)
By punkpeye on April 3, 2026.
mcp
The Hackers Who Tracked My Sleep Cycle
By punkpeye on March 26, 2026.
security

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/vishwajeetdabholkar/markdown-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server