M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
COMPARE UP TO 4 SERVERS

Compare MCP servers

Compare scores, permissions, risks, and fit in one decision-focused table.

DimensionJina AI Remote MCP ServerJina AI's official remote MCP server for web reading, search, and embeddings/rerankingFirecrawl MCP ServerFirecrawl's official MCP server for web search, scraping, and structured extraction for AI agents
FMRS64 / 100 · C75 / 100 · B
Reliability7 / 2013 / 20
Security and permissions13 / 2012 / 20
Maintenance12 / 2016 / 20
Documentation18 / 2017 / 20
Setup experience14 / 2017 / 20
Best for
  • Developers who want to give an LLM real-time web search and reading capability
  • Researchers doing academic literature review across arXiv and SSRN
  • Teams needing semantic deduplication or classification of text/image sets
  • Users with a Jina API key who want one remote MCP server usable across multiple clients
  • Scenarios needing clean, structured web context fed to an AI assistant
  • Users already on the Firecrawl platform who want to call its scraping capability directly via MCP
Not for
  • Fully offline or air-gapped environments with no outbound network access
  • Users unwilling to obtain a Jina API key, since most tools require one for full access and unauthenticated use is rate-limited
  • Latency-sensitive workflows, since tools like search_web_deep can take 2-20 seconds per call
  • Simple static-page scraping where you don't want to depend on a third-party API and incur call costs
  • Sites that explicitly disallow automated access (robots.txt)
Required permissions
  • An optional or required Jina API key (Bearer token) for accessing the Reader/Embeddings/Reranker APIs and higher rate limits
  • Outbound network access from the server to target URLs, search engines, and Jina's cloud APIs
  • Requires a Firecrawl API key to call; cost and quota are governed by the Firecrawl account
  • firecrawl_agent/firecrawl_interact perform automated browser interaction, which may trigger login or form-submission flows on the target site
Risks and side effects
  • This is a hosted remote server at mcp.jina.ai — queries and fetched page content pass through Jina's cloud infrastructure
  • API keys stored in plaintext config files or environment variables can be exposed if configs are shared or leaked
  • Rate limits apply to tools used without an API key, which can bottleneck heavy usage
  • Some tools (e.g. search_web_deep) have noticeably longer response times, which can affect interactive workflows
  • Bulk crawl/map tools can generate significant request volume against a target site — respect the site's rate limits and terms of service
  • firecrawl_agent's interactive action chain is longer — define task boundaries clearly before running it to avoid accidentally triggering actions on a sensitive site
Supported clientsCursor, LM Studio, Claude Code, Claude Desktop, Codex, WindsurfClaude Desktop, VS Code, Cursor, Windsurf, Zed, Amp
Tools2211