Context7
Up-to-date docs for AI agents
作者提供的说明(英文原文)
Analyst-debate research on every public and private company. 8 tools: search, score, compare, more.
这三项永远不会合并成一个分数:意见、测量和人气是不同的东西。
我们还没有为它写通俗的介绍。下面是我们从注册中心读取到的事实。
选择你使用的 AI 应用,然后复制设置。
你需要
你用的是哪个应用?
在上方选择你的应用。我们会记住它,下次按钮会显示“安装到 …”。
适用于网页版、桌面版和手机版 Claude:
https://mcp.askcyborg.com/mcp在终端中运行这条命令
claude mcp add --transport http askcyborg https://mcp.askcyborg.com/mcp粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json
{
"mcpServers": {
"askcyborg": {
"url": "https://mcp.askcyborg.com/mcp"
}
}
}浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。
粘贴到项目中的 .vscode/mcp.json
{
"servers": {
"askcyborg": {
"type": "http",
"url": "https://mcp.askcyborg.com/mcp"
}
}
}浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。
ChatGPT 只能连接互联网上的服务器:
https://mcp.askcyborg.com/mcp菜单名称可能因套餐和版本不同而略有差异。
你的选择通过 Cookie 保存在这台设备上,无需账户。
这些设置是根据作者在注册中心的条目生成的。如果 README 的写法不同,请以 README 为准。
我们还没有读取这个服务器的工具列表;我们会轮流检查各个端点。
我们检查的是任何人都能看到的信息;我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月7日
注册名称已通过 GitHub 账户 Ask-Cyborg 验证,该账户拥有此仓库
我们还没有从 GitHub 读取这个仓库
我们还没有从 GitHub 读取这个仓库
尚未检查;我们会轮流检查各个端点
尚未检查;我们会轮流检查各个端点
尚未检查;我们会在检查端点时读取工具列表
与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时,星标数才计入。
它在 1 项适用的基础检查中通过了 1 项。我们没有阅读代码。在添加密钥或令牌之前,请先看看这个服务器要求哪些权限,并使用只具备所需权限的密钥。
注册中心没有列出必需的密钥,不过有些服务器会在首次使用时要求你登录。
它在作者的云端运行。通过互联网连接,无需安装,但你发送的内容会经过作者的服务器。
GitHub 仓库最后更新于–,最新版本(0.2.0)于6个月前进入注册中心。
支持 MCP 的应用,例如 Claude Desktop、Claude Code、Cursor 和 VS Code。它通过 URL 连接,所以接受远程 MCP 服务器的应用也可以使用,例如开启开发者模式的 ChatGPT。
同一分类中有:Context7, Codebase Memory, Exa
同一分类中其他作者的服务器,按星标数排序。
如何看卡片:★ GitHub 星标i三大支柱:评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i
Up-to-date docs for AI agents
Codebase knowledge graph for AI agents — 162 languages, sub-ms queries, 99% fewer tokens.
Connect AI agents to Exa for web search, content fetching, and multi-step research.
按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。
Official MCP server for [AskCyborg](https://askcyborg.com). Search and retrieve AI company research, analyst-debate audio briefings, and the proprietary Cyborg Score (1-10) — directly from Claude, Cursor, Cline, Continue, Windsurf, ChatGPT, and any other Model Context Protocol client.
AskCyborg stress tests every public and private company through intense analyst debate, focused on what makes or breaks the business decisions that ride on them — packaged as 30-page reports, audio briefings, and a proprietary Cyborg Score (1-10) calculated from hundreds of data points per company.
| Tool | Purpose |
|---|---|
search_companies | Find companies in AskCyborg's corpus by name or industry keyword |
get_company_report | Full report — executive summary, Cyborg Score, strategic profile, top insights, competitors, recent developments |
get_cyborg_score | Quick rating-only lookup (1-10 + rationale) |
compare_companies | Side-by-side comparison of 2–5 companies on Cyborg Score, industry, and strategic profile |
find_competitors | Named competitors with one-line strategic taglines |
search_by_industry | Discover companies in a sector, ranked by Cyborg Score |
get_recent_developments | Dated news, deals, leadership changes, financial events |
get_top_insights | Just the punchiest analyst-debate insights, faster than the full report |
Each tool returns a structured Markdown response with a direct link to the full 30-page research report + 15-minute analyst-debate audio briefing on askcyborg.com.
Add to your claude_desktop_config.json:
{
"mcpServers": { "askcyborg": { "transport": { "type": "http", "url": "https://mcp.askcyborg.com/mcp" } } } }
Restart Claude Desktop. The tools appear in the model's tool list automatically.
.cursor/mcp.json in your project root:
{
"mcpServers": { "askcyborg": { "url": "https://mcp.askcyborg.com/mcp" } } }
Same pattern — point the client's MCP config at https://mcp.askcyborg.com/mcp via the HTTP transport.
Add a custom action with the OpenAPI spec at https://mcp.askcyborg.com/openapi.json (coming in v0.2).
Find AI infrastructure companies:
Use AskCyborg to find five companies in the AI infrastructure space and tell me which has the highest Cyborg Score.
One-shot research:
Pull the AskCyborg report on Anthropic and summarize the bull case in three bullets.
Comparison for due diligence:
Compare OpenAI, Anthropic, and Mistral on AskCyborg's Cyborg Score and competitive positioning.
Watchlist screening:
For each of CRWD, ZS, S, NET, PANW — pull the Cyborg Score from AskCyborg and rank.
search_companies, get_company_report, get_cyborg_score, compare_companies)get_full_report, get_audio_briefing_url, watchlists)compare_companies extended to 10 entries with deeper diffget_competitive_landscape (full competitor graph, not just 3)get_recent_filings_summary for S-1, 10-K, 10-Qgit clone https://github.com/Ask-Cyborg/askcyborg-mcp.git
cd askcyborg-mcp npm install npm run dev # starts wrangler dev on http://localhost:8787
In a separate terminal:
curl -X POST http://localhost:8787/mcp \
-H 'Content-Type: application/json' \ -d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'
To deploy your own instance:
wrangler secret put SUPABASE_URL
wrangler secret put SUPABASE_ANON_KEY npm run deploy
AskCyborg covers millions of public and private companies across 195 countries. Every company gets a 30-page research report, a 15-minute analyst-debate podcast (where bull and bear AI analysts argue the investment case), and a Cyborg Score (1-10) synthesized from hundreds of data points.
Built for VCs, finance teams, corporate strategists, and operators whose decisions ride on understanding companies fast and from every angle.
MIT. Use this MCP server in any client. The underlying AskCyborg content (research reports, audio briefings, Cyborg Score) remains AskCyborg's intellectual property — please respect the rate limits and don't bulk-scrape.
Open an issue at https://github.com/Ask-Cyborg/askcyborg-mcp/issues. PRs welcome for new tools, client examples, and bug fixes.