mcp-scraper 0.3.15 → 0.3.16

package/docs/mcp-tool-manifest.generated.json

@@ -1,12 +1,12 @@
 {
-  "generatedAt": "2026-06-25T21:55:09.732Z",
+  "generatedAt": "2026-06-26T17:22:50.447Z",
   "counts": {
-    "main_stdio": 24,
+    "main_stdio": 25,
     "browser_agent_stdio": 20,
-    "combined_stdio": 44,
-    "hosted_http": 26,
+    "combined_stdio": 45,
+    "hosted_http": 27,
     "hosted_only": 2,
-    "unique_public": 46
+    "unique_public": 47
   },
   "surfaces": {
     "main_stdio": [
@@ -15,6 +15,7 @@
       "extract_url",
       "map_site_urls",
       "extract_site",
+      "audit_site",
       "youtube_harvest",
       "youtube_transcribe",
       "facebook_page_intel",
@@ -63,6 +64,7 @@
       "extract_url",
       "map_site_urls",
       "extract_site",
+      "audit_site",
       "youtube_harvest",
       "youtube_transcribe",
       "facebook_page_intel",
@@ -109,6 +111,7 @@
       "extract_url",
       "map_site_urls",
       "extract_site",
+      "audit_site",
       "youtube_harvest",
       "youtube_transcribe",
       "facebook_page_intel",
@@ -141,6 +144,7 @@
       "extract_url",
       "map_site_urls",
       "extract_site",
+      "audit_site",
       "youtube_harvest",
       "youtube_transcribe",
       "facebook_page_intel",
@@ -322,18 +326,49 @@
       "hasStructuredContent": true,
       "writesLocalFiles": true,
       "symptomTriggers": [
-        "the user wants a multi-page site audit, content crawl, competitor crawl, or schema inventory"
+        "the user wants the page content/text of multiple pages on a site (a content crawl or scrape)"
       ],
       "samplePrompts": [
-        "Extract this site for an SEO audit",
-        "Crawl the top pages of this competitor",
-        "Get titles and schema across the site"
+        "Scrape the content of these pages",
+        "Crawl the top pages of this competitor for their copy",
+        "Get the page content across the site"
       ],
       "formatExamples": [
         "url=\"https://example.com\"",
         "maxPages=50 for a normal crawl, up to maxPages=10000 for a full-site bulk scrape (crawls over 25 pages switch to folder mode: each page saved as its own file, response returns a summary plus the folder path)"
       ],
-      "negativeSpace": "Use map_site_urls first when the user needs URL selection; use extract_url for one page.",
+      "negativeSpace": "Returns content only. Use audit_site for a technical SEO audit (issues, link graph, indexability, image weights); use map_site_urls first when the user needs URL selection; use extract_url for one page.",
+      "producesHandlesFor": [
+        "saved-report"
+      ],
+      "consumesHandles": [
+        "public-site-url"
+      ]
+    },
+    {
+      "name": "audit_site",
+      "surfaces": [
+        "main_stdio",
+        "combined_stdio",
+        "hosted_http"
+      ],
+      "tokenBudgetClass": "core",
+      "hasOutputSchema": true,
+      "hasStructuredContent": true,
+      "writesLocalFiles": true,
+      "symptomTriggers": [
+        "the user wants a technical SEO audit, site health check, or Screaming-Frog-style crawl with issues, internal link graph, indexability, heading breakdown, and image sizes/formats"
+      ],
+      "samplePrompts": [
+        "Run a technical SEO audit on this site",
+        "Do a Screaming Frog style crawl of example.com",
+        "Audit this site for SEO issues and oversized images"
+      ],
+      "formatExamples": [
+        "url=\"https://example.com\"",
+        "maxPages=50 for a normal audit, up to 10000 for a full-site audit; always writes a folder of analysis files (report.md, issues.json, pages.jsonl, links.jsonl, link-metrics.jsonl, images.jsonl) plus per-page content and returns a summary plus the folder path"
+      ],
+      "negativeSpace": "Use extract_site when the user only wants page content; use map_site_urls to just list URLs.",
       "producesHandlesFor": [
         "saved-report"
       ],

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "mcp-scraper",
-  "version": "0.3.15",
+  "version": "0.3.16",
   "description": "MCP server for MCP Scraper web intelligence tools",
   "type": "module",
   "main": "./dist/index.cjs",

package/dist/chunk-2MX5WOII.js

@@ -1,7 +0,0 @@
-// src/version.ts
-var PACKAGE_VERSION = "0.3.15";
-
-export {
-  PACKAGE_VERSION
-};
-//# sourceMappingURL=chunk-2MX5WOII.js.map

package/dist/chunk-2MX5WOII.js.map

@@ -1 +0,0 @@
-{"version":3,"sources":["../src/version.ts"],"sourcesContent":["export const PACKAGE_VERSION = '0.3.15'\n"],"mappings":";AAAO,IAAM,kBAAkB;","names":[]}

package/dist/chunk-7Y6JDBML.js.map

@@ -1 +0,0 @@
-{"version":3,"sources":["../src/install-terminal.ts"],"sourcesContent":["export interface InstallTerminalOptions {\n  color?: boolean\n  version: string\n  apiKeyConfigured?: boolean\n}\n\ntype Tone = 'cyan' | 'lime' | 'amber' | 'red' | 'muted' | 'bold'\n\nconst CODES: Record<Tone | 'reset', string> = {\n  reset: '\\x1b[0m',\n  cyan: '\\x1b[36m',\n  lime: '\\x1b[32m',\n  amber: '\\x1b[33m',\n  red: '\\x1b[31m',\n  muted: '\\x1b[90m',\n  bold: '\\x1b[1m',\n}\n\nfunction colorize(value: string, tone: Tone, enabled: boolean): string {\n  if (!enabled) return value\n  return `${CODES[tone]}${value}${CODES.reset}`\n}\n\nfunction toolRow(label: string, tools: string[], enabled: boolean): string {\n  const padded = label.padEnd(9, ' ')\n  return `  ${colorize(padded, 'muted', enabled)} ${tools.join(colorize('  .  ', 'muted', enabled))}`\n}\n\nexport function renderInstallTerminal(options: InstallTerminalOptions): string {\n  const color = options.color ?? true\n  const apiKeyValue = options.apiKeyConfigured ? '$MCP_SCRAPER_API_KEY' : 'sk_live_your_key'\n  const ascii = String.raw`\n __  __  ____ ____ \n|  \\/  |/ ___|  _ \\\n| |\\/| | |   | |_) |\n| |  | | |___|  __/\n|_|  |_|\\____|_|\n\n ____   ____ ____      _    ____  _____ ____\n/ ___| / ___|  _ \\    / \\  |  _ \\| ____|  _ \\\n\\___ \\| |   | |_) |  / _ \\ | |_) |  _| | |_) |\n ___) | |___|  _ <  / ___ \\|  __/| |___|  _ <\n|____/ \\____|_| \\_\\/_/   \\_\\_|   |_____|_| \\_\\\n`\n\n  const claudeCommand = [\n    `MCP_SCRAPER_API_KEY=${apiKeyValue} npx -y -p mcp-scraper@latest \\\\`,\n    '  mcp-scraper-cli agent install claude --apply',\n  ].join('\\n')\n\n  const codexConfig = [\n    '[mcp_servers.mcp-scraper]',\n    'command = \"npx\"',\n    'args = [\"-y\", \"-p\", \"mcp-scraper@latest\", \"mcp-scraper-combined\"]',\n    `env = { MCP_SCRAPER_API_KEY = \"${apiKeyValue}\" }`,\n  ].join('\\n')\n\n  return [\n    colorize(`mcp-scraper v${options.version}`, 'bold', color),\n    colorize('> mcp-scraper-install', 'muted', color),\n    colorize(ascii, 'amber', color),\n    `${colorize('MCP Scraper Agent', 'cyan', color)}  .  v${options.version}  .  mcpscraper.dev`,\n    '1/1 install surfaces ready',\n    colorize('Newest: saved hosted browser profiles for AI visibility. Share the watch_url, user signs in, then verify with browser_profile_status. Browser sessions are direct/no-proxy by default.', 'lime', color),\n    '',\n    `${colorize('Tools', 'cyan', color)}  ${colorize('(44 MCP tools)', 'muted', color)}`,\n    toolRow('search', ['harvest_paa', 'search_serp', 'maps_search', 'maps_place_intel'], color),\n    toolRow('extract', ['extract_url', 'map_site_urls', 'extract_site', 'directory_workflow'], color),\n    toolRow('build', ['rank_tracker_workflow', 'cron plan', 'database prompt'], color),\n    toolRow('media', ['youtube_harvest', 'youtube_transcribe', 'facebook_ad_search', 'facebook_page_intel', 'facebook_ad_transcribe', 'facebook_video_transcribe', 'instagram_profile_content', 'instagram_media_download'], color),\n    toolRow('browser', ['browser_open', 'browser_profile_onboard', 'browser_profile_status', 'browser_close', 'browser_screenshot', 'browser_read', 'browser_locate', 'browser_replay_mark', 'browser_replay_annotate'], color),\n    toolRow('account', ['credits_info', 'reports', 'MCP resources'], color),\n    `${colorize('Workflows', 'cyan', color)}  ${colorize('(MCP + CLI + API)', 'muted', color)}`,\n    toolRow('route', ['workflow_list', 'workflow_suggest', 'workflow_run', 'workflow_step', 'workflow_status', 'workflow_artifact_read'], color),\n    toolRow('seo', ['directory', 'agent-packet', 'competitive audit', 'map/serp comparison', 'PAA/AIO briefs', 'scheduled runs'], color),\n    '',\n    colorize('Usage tips:', 'amber', color),\n    'Run mcp-scraper-install for this visible card. Run mcp-scraper-cli for setup utilities and subcommands.',\n    'Run mcp-scraper-combined in a human terminal to print this card; MCP clients get the same command as a silent stdio server.',\n    'Explicit card command: npx -y -p mcp-scraper@latest mcp-scraper-install',\n    'Customer auth setup: run browser_profile_onboard, send the watch_url, let the user sign in, then call browser_profile_status until AUTHENTICATED.',\n    'Reconnect setup: if a saved profile shows login controls later, run browser_profile_onboard again with the same profile name.',\n    'Start with workflow_suggest for broad jobs like market analysis, ICP research, CRO audits, brand briefs, content gaps, and AI visibility.',\n    'For MCP clients, use mcp-scraper-combined so one install can mix SERP, Maps, browser, reports, and saved MCP resources.',\n    'If you hit the concurrency limit, add an extra slot for $5/month with mcp-scraper-cli billing concurrency checkout.',\n    '',\n    `${colorize('Ready.', 'lime', color)} Install the combined MCP server with one command:`,\n    '',\n    colorize('Setup doctor', 'amber', color),\n    'npx -y -p mcp-scraper@latest mcp-scraper-cli doctor',\n    '',\n    colorize('Hosted profile setup', 'amber', color),\n    'In your MCP client, call browser_profile_onboard with email=\"seo@example.com\" and domain=\"chatgpt.com\".',\n    'Give the returned watch_url to the user. After they sign in, call browser_profile_status, then browser_open with the returned profile.',\n    '',\n    colorize('Claude Code one-command setup', 'amber', color),\n    claudeCommand,\n    'Then fully exit Claude Code and open a new Claude terminal. Check with: claude mcp list',\n    '',\n    colorize('Codex config', 'amber', color),\n    codexConfig,\n    '',\n    colorize('Claude Desktop Extension', 'amber', color),\n    'Download: https://mcpscraper.dev/downloads/mcp-scraper.mcpb',\n    '',\n    colorize('Safety note:', 'muted', color),\n    'mcp-scraper-combined prints this card only when stdin/stdout are an interactive TTY. In MCP clients it writes only JSON-RPC to stdout.',\n    'Use --stdio or MCP_SCRAPER_FORCE_STDIO=1 to force server mode from a terminal.',\n    '',\n  ].join('\\n')\n}\n"],"mappings":";AAQA,IAAM,QAAwC;AAAA,EAC5C,OAAO;AAAA,EACP,MAAM;AAAA,EACN,MAAM;AAAA,EACN,OAAO;AAAA,EACP,KAAK;AAAA,EACL,OAAO;AAAA,EACP,MAAM;AACR;AAEA,SAAS,SAAS,OAAe,MAAY,SAA0B;AACrE,MAAI,CAAC,QAAS,QAAO;AACrB,SAAO,GAAG,MAAM,IAAI,CAAC,GAAG,KAAK,GAAG,MAAM,KAAK;AAC7C;AAEA,SAAS,QAAQ,OAAe,OAAiB,SAA0B;AACzE,QAAM,SAAS,MAAM,OAAO,GAAG,GAAG;AAClC,SAAO,KAAK,SAAS,QAAQ,SAAS,OAAO,CAAC,IAAI,MAAM,KAAK,SAAS,SAAS,SAAS,OAAO,CAAC,CAAC;AACnG;AAEO,SAAS,sBAAsB,SAAyC;AAC7E,QAAM,QAAQ,QAAQ,SAAS;AAC/B,QAAM,cAAc,QAAQ,mBAAmB,yBAAyB;AACxE,QAAM,QAAQ,OAAO;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAcrB,QAAM,gBAAgB;AAAA,IACpB,uBAAuB,WAAW;AAAA,IAClC;AAAA,EACF,EAAE,KAAK,IAAI;AAEX,QAAM,cAAc;AAAA,IAClB;AAAA,IACA;AAAA,IACA;AAAA,IACA,kCAAkC,WAAW;AAAA,EAC/C,EAAE,KAAK,IAAI;AAEX,SAAO;AAAA,IACL,SAAS,gBAAgB,QAAQ,OAAO,IAAI,QAAQ,KAAK;AAAA,IACzD,SAAS,yBAAyB,SAAS,KAAK;AAAA,IAChD,SAAS,OAAO,SAAS,KAAK;AAAA,IAC9B,GAAG,SAAS,qBAAqB,QAAQ,KAAK,CAAC,SAAS,QAAQ,OAAO;AAAA,IACvE;AAAA,IACA,SAAS,0LAA0L,QAAQ,KAAK;AAAA,IAChN;AAAA,IACA,GAAG,SAAS,SAAS,QAAQ,KAAK,CAAC,KAAK,SAAS,kBAAkB,SAAS,KAAK,CAAC;AAAA,IAClF,QAAQ,UAAU,CAAC,eAAe,eAAe,eAAe,kBAAkB,GAAG,KAAK;AAAA,IAC1F,QAAQ,WAAW,CAAC,eAAe,iBAAiB,gBAAgB,oBAAoB,GAAG,KAAK;AAAA,IAChG,QAAQ,SAAS,CAAC,yBAAyB,aAAa,iBAAiB,GAAG,KAAK;AAAA,IACjF,QAAQ,SAAS,CAAC,mBAAmB,sBAAsB,sBAAsB,uBAAuB,0BAA0B,6BAA6B,6BAA6B,0BAA0B,GAAG,KAAK;AAAA,IAC9N,QAAQ,WAAW,CAAC,gBAAgB,2BAA2B,0BAA0B,iBAAiB,sBAAsB,gBAAgB,kBAAkB,uBAAuB,yBAAyB,GAAG,KAAK;AAAA,IAC1N,QAAQ,WAAW,CAAC,gBAAgB,WAAW,eAAe,GAAG,KAAK;AAAA,IACtE,GAAG,SAAS,aAAa,QAAQ,KAAK,CAAC,KAAK,SAAS,qBAAqB,SAAS,KAAK,CAAC;AAAA,IACzF,QAAQ,SAAS,CAAC,iBAAiB,oBAAoB,gBAAgB,iBAAiB,mBAAmB,wBAAwB,GAAG,KAAK;AAAA,IAC3I,QAAQ,OAAO,CAAC,aAAa,gBAAgB,qBAAqB,uBAAuB,kBAAkB,gBAAgB,GAAG,KAAK;AAAA,IACnI;AAAA,IACA,SAAS,eAAe,SAAS,KAAK;AAAA,IACtC;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,IACA,GAAG,SAAS,UAAU,QAAQ,KAAK,CAAC;AAAA,IACpC;AAAA,IACA,SAAS,gBAAgB,SAAS,KAAK;AAAA,IACvC;AAAA,IACA;AAAA,IACA,SAAS,wBAAwB,SAAS,KAAK;AAAA,IAC/C;AAAA,IACA;AAAA,IACA;AAAA,IACA,SAAS,iCAAiC,SAAS,KAAK;AAAA,IACxD;AAAA,IACA;AAAA,IACA;AAAA,IACA,SAAS,gBAAgB,SAAS,KAAK;AAAA,IACvC;AAAA,IACA;AAAA,IACA,SAAS,4BAA4B,SAAS,KAAK;AAAA,IACnD;AAAA,IACA;AAAA,IACA,SAAS,gBAAgB,SAAS,KAAK;AAAA,IACvC;AAAA,IACA;AAAA,IACA;AAAA,EACF,EAAE,KAAK,IAAI;AACb;","names":[]}