"Using Claude AI on a desktop for web searches" matching MCP tools:

• post_anthropic_websearch_searchA

  AIsa Web Search & ResearchA

  Ask a question and get an answer that Claude wrote after searching the live web. Send a normal Messages request — a `messages` array plus `max_tokens` — and the endpoint injects a fixed, server-pinned model and the `web_search` tool for you; you cannot override the model or add tools, which keeps cost bounded. Claude decides when to search (up to `max_uses` searches, default 5), reads the results, and answers with inline citations. The response is a standard Anthropic Messages object: `content[]` contains `server_tool_use` (the queries issued), `web_search_tool_result` (the sources found) and `text` blocks (the answer with `citations`), and `usage.server_tool_use.web_search_requests` reports how many searches were billed. a failed search (HTTP 200 `web_search_tool_result_error`) is not billed. Use this when you want a written, cited answer grounded in current web content — for open-web research that returns ranked links and page text in one call use [`post_tavily_search`](/api-reference/search/post_tavily-search) instead, and for the OpenAI-model equivalent see [`post_openai_websearch_search`](/api-reference/search/post_openai-websearch-search). <!-- aisa-pricing:start --> Usage-based billing; Default request estimate: $0.3 USD (not a fixed price). Account-specific pricing and final settlement may differ. <!-- aisa-pricing:end -->

  Connector

  Destructive

  OAuth

• aeo_run_ai_answersA

  Manifold MCPA

  What AI engines answer for a prompt, with the brands they mention and the sources they cite. Slow: returns a task_id, then read the result with get_task. Use when: "does ChatGPT recommend us for X", "who do AI engines cite for X", an AI visibility baseline for a brand across engines. Not for: which prompts people ask, use aeo_search_prompts. Google's ten blue links, use seo_get_serp (ai_overview: true adds the overview to it at 1 credit). Whether a site is ready for AI crawlers, use aeo_get_site_readiness. Returns: Task. task_id now; get_task returns rows[] per prompt and engine (answer, mentions[] (brand, mentioned, position, cited), citations[] (position, url, domain, title), search_results[] (pages retrieved, cited or not), search_queries[] (the web searches the engine ran), named_brands[] (brands ChatGPT tagged in the answer, not only yours)) and errors[] for cells that failed. answer null = the engine showed no AI answer. search_results[] and named_brands[] are filled for chatgpt only, search_queries[] for chatgpt and claude. named_brands[] drops a tag ChatGPT gives as a description instead of the brand, so it can miss brands; read the answer for a full list. chatgpt and gemini are what their apps show a person, model chosen by the app; claude and perplexity come from their model APIs. Answers are live and non-deterministic; results are kept 30 days and never cached. Cost: per prompt, 15 credits for claude, 6 for perplexity, 2 each for chatgpt, gemini, ai_overview and ai_mode. Default 1 prompt on the five default engines = 27 credits. A cell that fails is not charged. Example: {"prompts": ["best crm for startups"], "brands": ["HubSpot", "pipedrive.com"]}

  ConnectorAPI key

• get_page_performanceA

  rankparse-mcpA

  Google PageSpeed Insights report for a URL: Lighthouse performance/accessibility/SEO scores and Core Web Vitals (LCP, CLS, INP, FCP, TTFB), with lab and field (CrUX) data. Defaults to mobile; pass strategy=desktop for the desktop profile. Subject to a per-user daily cap (50/day) and a service-wide daily cap; successful responses are cached for 24h.

  ConnectorNo auth

• list_runtimesA

  MCPExplorerA

  List the runtimes generate_runtime_config supports (Claude Desktop, Cursor, VS Code, agent frameworks, …), with each one's config path. Enumerate these instead of guessing runtime slugs.

  ConnectorNo auth

• brave_web_searchA

  braveA

  Performs web searches using the Brave Search API and returns comprehensive search results with rich metadata. To chain into local-POI enrichment, pass `result_filter=locations` and feed the resulting `locations.results[].id` values into `brave_local_search`. To chain into the AI summarizer, pass `summary=true` and feed the returned `summarizer.key` into `brave_summarizer`.

  ConnectorOAuth

• upload_mediaA

  SendItB

  FOR CLAUDE DESKTOP ONLY (with filesystem access). For Claude.ai/web: Use create_upload_session instead - it provides a browser upload link. Upload local media to cloud storage, returning a public HTTPS URL. WHEN TO USE: • Instagram, LinkedIn, Threads, X: REQUIRED for local files before calling publish_content • TikTok: NOT NEEDED - pass local path directly to publish_content SUPPORTED FORMATS: • Images: jpg, png, gif, webp (max 10MB) • Videos: mp4, mov, webm (max 100MB) Returns { url: 'https://...' } for use in publish_content mediaUrl parameter.

  ConnectorNo auth

• get_setup_instructionsA

  FundingLandscapeA

  Return MCP setup instructions for Claude Desktop and other clients. Does not count toward your monthly searches.

  ConnectorNo auth

• search_webA

  Verify-apiA

  Execute live web searches using multi-engine chain (TinyFish, DuckDuckGo, Jina) without monthly API subscriptions. Returns fresh source URLs, titles, and snippets. Requires x402 micropayment (0.005 USDC on Base). When to use: Real-time web browsing and information retrieval for AI agents. When NOT to use: Do NOT use for deep recursive crawling of entire sites. Parameters: - `query` (string, required): Search query (3-300 chars). - `limit` (integer, optional, default 5): Maximum number of search results to return (1-10).

  ConnectorNo auth

• wikimedia-analytics.pageviews.topA

  apibaseB

  List the most-viewed articles on a Wikimedia project for a single day, or for a full month with day="all-days". Returns each article title, view count, and rank. Filter by access method (desktop, mobile-web, mobile-app, or all-access). Data: Wikimedia Analytics REST API (wikimedia.org/api/rest_v1), no auth required.

  ConnectorNo auth

• search_webA

  verify-apiA

  Execute live web searches using multi-engine chain (TinyFish, DuckDuckGo, Jina) without monthly API subscriptions. Returns fresh source URLs, titles, and snippets. Requires x402 micropayment (0.005 USDC on Base). When to use: Real-time web browsing and information retrieval for AI agents. When NOT to use: Do NOT use for deep recursive crawling of entire sites. Parameters: - `query` (string, required): Search query (3-300 chars). - `limit` (integer, optional, default 5): Maximum number of search results to return (1-10).

  ConnectorNo auth

• fast_searchA

  ScrapingBeeA

  Search the web using ScrapingBee's Fast Search API and return structured results. This is the PRIMARY tool for web searches — it returns clean organic results and top stories in under one second at a lower credit cost. USE THIS TOOL FIRST for any general web search. Only fall back to get_google_search_results if you need specialized search types such as news, maps, Google Lens, shopping, image search, or Google AI mode. Scope: one query per call. To run many queries in one pass, or to write results to disk instead of into the conversation, use the ScrapingBee CLI — `scrapingbee <command> --input-file queries.txt --output-dir results`.

  ConnectorAPI key

• fast_searchA

  ScrapingBeeA

  Search the web using ScrapingBee's Fast Search API and return structured results. This is the PRIMARY tool for web searches — it returns clean organic results and top stories in under one second at a lower credit cost. USE THIS TOOL FIRST for any general web search. Only fall back to get_google_search_results if you need specialized search types such as news, maps, Google Lens, shopping, image search, or Google AI mode. Scope: one query per call. To run many queries in one pass, or to write results to disk instead of into the conversation, use the ScrapingBee CLI — `scrapingbee <command> --input-file queries.txt --output-dir results`.

  ConnectorNo auth

• ai_answersA

  FenlightB

  Put a buyer's question to an AI assistant that searches the live web, and see whether a brand and its competitors are named in the answer, in what rank, with the citations and cited domains behind it. The assistant is ChatGPT unless `engine` names claude, perplexity or gemini; an assistant the server has no key for answers not_configured and costs nothing. `samples` (1 to 5, default 1) asks the same question more than once, for a mention rate. Returns metrics per assistant, not the raw answers. Charged per call to the assistant at the price list in credits (see llms.txt), the same as POST /v1/ai/answers: the worst case is held first, only the calls made are charged, and a call where the assistant failed is charged nothing.

  ConnectorNo auth

• configure_clientsA

  Local MCPA

  Adds Local MCP to the config of the MCP-capable AI apps installed on this Mac, from LMCP's client catalog (Claude Desktop, Claude Code, Cursor, Windsurf, VS Code, Zed, Codex / ChatGPT desktop, Jan, LM Studio, Warp, Antigravity, Cline, Roo and more) — no manual JSON editing. Read-only PREVIEW unless confirm:true. Optionally pass a single `client` id. Reports per client whether it was set up, was already set up, can't be set up automatically (and why), or isn't installed, plus the step the user must take for the change to apply (restart the app, or open a new session). Pair with list_missing_permissions for fully agent-driven setup.

  ConnectorOAuth

• verify_ai_claimA

  verify-apiA

  Verify whether an AI model, API provider, pricing claim, or infrastructure assertion is true today using live web evidence. Returns deterministic verdicts (TRUE, FALSE, PARTIALLY_TRUE, UNREACHABLE, UNVERIFIED) with cited sources and confidence scores. Requires x402 micropayment (0.01 USDC on Base). When to use: Fact-checking an AI provider's claims, pricing, or model availability. When NOT to use: Do NOT use for general open-ended web search, coding assistance, or non-AI claim verification. Parameters: - `query` (string, required): The exact claim or assertion to verify (5-500 chars), e.g. 'Is GLM-5.3 Flash free on ZenMux?'. - `depth` (string, optional, default 'standard'): Verification depth. 'standard' executes fast single-pass search (0.01 USDC); 'deep' conducts multi-source cross-examination (0.03 USDC).

  ConnectorNo auth

• vc_marketplace_searchA

  Vibes-Coded Agent Security and Commerce ToolsA

  Search the Vibes-Coded marketplace for AI-agent tools and API outcomes. Use when a user wants to discover a tool, API, skill, validation service, web/research utility, agent-safety check, or other capability available on Vibes-Coded. Search by the user's actual task and return relevant listings. This tool only searches Vibes-Coded's bounded marketplace catalog. It does not execute a paid outcome, change user data, or search the general web. Prefer this before `pay` when the user has not already chosen a specific Vibes-Coded outcome.

  ConnectorNo auth

• thinkneo_registry_installA

  ThinkNEO Control PlaneC

  Get installation config for an MCP server from the ThinkNEO Marketplace. Returns ready-to-use JSON config for Claude Desktop, Cursor, Windsurf, or custom clients. Tracks the download. No authentication required.

  ConnectorNo auth

• loginA

  RipostiqA

  Claude Code plugin login only (auth_token flow). On the web/desktop connector, identity comes from the connector's OAuth approval — do NOT ask the user to type their email/password into the chat; direct them to the connector's 'Log in with Ripostiq' or ripostiq.com instead.

  ConnectorNo auth

• ai_hub_searchA

  EnpitechA

  Searches the Enpitech AI Hub, Enpitech's curated catalog of Claude Code skills, MCP servers and AI tools, and agentic workflows for frontend teams, each one used by Enpitech engineers in production. Use it when the user wants a skill, MCP server, AI tool or workflow for a frontend task (code review, Figma to code, testing, context files, generative UI and so on), or asks what the AI Hub has. Returns matching items with a one-line take, an overview, the install command when there is one, and a link to the item's page on enpitech.dev. Read-only: it searches a published catalog and sends nothing anywhere.

  ConnectorNo auth

• search_guidesA

  VibeRaven GuidesA

  Use this when the user has a launch question about an AI-built app on Vercel + Supabase, such as Supabase RLS, env vars on Vercel, a Stripe webhook after deploy or a client handoff. Searches the launch guides published on viberaven.dev by keyword and returns up to 5 matches, each with its title, URL and the short answer from the page. It only searches those published pages: it does not browse the web or read the user's project. No match returns an empty list.

  ConnectorNo auth

• get_install_configA

  FastDropA

  Ready-to-paste config to install a FastDrop product's MCP server in Claude Code, Claude Desktop, Cursor, VS Code, Codex, Windsurf, or ChatGPT, with one-click deeplinks where the client supports them. Use when the user asks how to install, connect, or add a server found on FastDrop. Secrets are always <PLACEHOLDER> values.

  ConnectorNo auth