generated: '2026-10-07' method: probed status: published source: https://s1.dev/docs/integrations/mcp; https://s1.dev/.well-known/s1-dev-mcp-server-card.json; anonymous POST tools/list to https://mcp.search1api.com/mcp (401, OAuth challenge) on 2026-10-07 description: Search1API operates an official hosted MCP server (Streamable HTTP) at https://mcp.search1api.com/mcp exposing five tools (search, news, crawl, sitemap, trending) over OAuth 2.1 (PKCE + dynamic client registration via clerk.s1.dev) or a Search1API API key. The same server ships as the npm package search1api-mcp for local stdio use. Tool names and input schemas below are the ones the provider PUBLISHES in its MCP server card and docs table; an anonymous tools/list was refused with 401 (0.6.1 restored authentication on every /mcp request), so no live tool listing was observed. server: name: search1api transport: streamable-http url: https://mcp.search1api.com/mcp auth: 'oauth2 or api-key (Authorization: Bearer)' protocol_revision_claimed: 2026-07-28 (docs); server card protocolVersion 2025-06-18 protected_resource: https://mcp.search1api.com/.well-known/oauth-protected-resource authorization_server: https://clerk.s1.dev server_card: well-known/s1-dev-mcp-server-card.json version: 0.6.1 local: name: search1api-mcp transport: stdio install: npx -y search1api-mcp env: - SEARCH1API_KEY source: https://github.com/superagents-lab/search1api-mcp registry: https://www.npmjs.com/package/search1api-mcp version: 0.6.1 published: '2026-09-03' deployment: mode: both endpoint: https://mcp.search1api.com/mcp install: npx -y search1api-mcp package: https://www.npmjs.com/package/search1api-mcp auth: oauth verified: probed probe: gated probe_detail: POST tools/list without credentials -> HTTP 401 {"jsonrpc":"2.0","error":{"code":-32001,"message":"Authentication is required. Use OAuth 2.1 discovery or provide a Search1API API key as a Bearer token."}} checked: '2026-10-07' tools: - name: search description: Search the live public web when the user needs current information, sources, or research. Returns citable results with id, title, URL, and text. Each result includes a URL that can be read in full. inputs: - query - max_results - search_service - crawl_results - include_sites - exclude_sites - time_range required: - query source: https://s1.dev/.well-known/s1-dev-mcp-server-card.json - name: news description: Search current news when the user asks about recent events, announcements, or coverage. Returns citable article results with title, URL, snippet, and optional extracted text. inputs: - query - max_results - search_service - crawl_results - include_sites - exclude_sites - time_range required: - query source: https://s1.dev/.well-known/s1-dev-mcp-server-card.json - name: crawl description: Extract the readable title and full text from a public URL supplied by the user or returned by search. inputs: - url required: - url source: https://s1.dev/.well-known/s1-dev-mcp-server-card.json - name: sitemap description: Discover related public links from a page or domain when the user wants to explore a site's structure or available pages. inputs: - url required: - url source: https://s1.dev/.well-known/s1-dev-mcp-server-card.json - name: trending description: List currently trending repositories or stories from GitHub or Hacker News when the user asks what is popular now. inputs: - search_service - max_results required: - search_service source: https://s1.dev/.well-known/s1-dev-mcp-server-card.json tool_evidence: l3: table evidence_url: https://s1.dev/docs/integrations/mcp note: Five-row Tool/What it does/Required table on the docs page, matching the server card; names are documented, not observed from a live tools/list. directories: - https://claude.ai/directory/search1api - https://smithery.ai (listed per docs) - https://cursor.directory (Agent Plugins manifest plugin.json in repo) secondary_servers: - name: Search1API Blog MCP url: https://blog.s1.dev/mcp transport: streamable-http verified: probed note: Anonymous tools/list answered 200 with four read-only blog content tools; its protected-resource document names https://blog.s1.dev/_emdash as authorization server for write scopes. A content server for the blog, not the search API. tools: - name: search_posts description: Search Search1API Blog posts by keyword. Returns matching posts with excerpts and URLs. inputs: - query - limit - name: list_posts description: List Search1API Blog posts in reverse chronological order. Optionally filter by tag slug and paginate with nextCursor. inputs: - limit - cursor - tag - name: get_post description: Get a single Search1API Blog post by slug, including plain-text body content. inputs: - slug - name: list_tags description: List all tags used on the Search1API Blog. inputs: []