Robin: AI-Powered Dark Web OSINT
Dark web OSINT over Tor: search onion engines, scrape pages, report with your own model.
作者提供的说明(英文原文)
378k+ ERC-8004 agents and their security verifications. Two tools free, the rest USDC over x402.
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。
Authorization标为机密的值是你账户的钥匙。千万不要分享给别人,也不要粘贴到聊天里。
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 …”。
这个地址需要在请求的 header 中放入密钥,而添加连接器的菜单无法设置。请改用 Claude Code、Cursor 或 VS Code。
在终端中运行这条命令
claude mcp add --transport http erc-8126-scan https://erc8126scan.ai/api/mcp --header 'Authorization: your-value'运行前请先填写空着的值。
粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json
{
"mcpServers": {
"erc-8126-scan": {
"url": "https://erc8126scan.ai/api/mcp",
"headers": {
"Authorization": "your-value"
}
}
}
}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": {
"erc-8126-scan": {
"type": "http",
"url": "https://erc8126scan.ai/api/mcp",
"headers": {
"Authorization": "your-value"
}
}
}
}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。
这个地址需要在请求的 header 中放入密钥,而添加连接器的菜单无法设置。请改用 Claude Code、Cursor 或 VS Code。
你的选择通过 Cookie 保存在这台设备上,无需账户。
Authorization 机密 HTTP header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。
工具名称和描述与服务器在 2026年11月10日 报告的完全一致。这些内容由作者撰写,不是我们写的;我们以纯文本形式显示。
list_servicesFree. The catalogue: every tool on this server, what it returns, what it costs per call, and how to read the risk fields. Call this first — it is authoritative about pricing, needs no payment, key or wallet, and answers most questions about what this server can do without spendin
get_index_statsFree. How much of the ERC-8004 agent population this index covers: total agents synced from the on-chain Identity Registry, and how many carry verification results. Use it to judge whether the index is likely to know about the agent you care about before paying to look it up.
search_agentsSearch and filter scanned ERC-8004 agents by chain, risk score range, and verification status. Returns a paginated list with per-dimension scores (ETV, MCV, SCV, WAV, WV) and an overall risk score. Pass `wallet` or `id` to look up a single agent directly. IMPORTANT: read `risk_le
get_agentFetch the full detail record for a single agent by UUID, including verification state and metadata. Judge the agent by `risk_level`, not the raw overall_risk_score: 'not_verified' means nobody has paid to assess it yet and scores 100 by presumption, which is not a finding against
get_agent_activityFetch on-chain activity and liveness signals for one agent, including whether it is currently active and when it was last checked. Costs $0.002 USDC per call when paying with x402.
get_agent_reportFetch the full security assessment report for one agent: per-dimension findings, evidence and recommendations. Returns data only for agents owned by the caller — the wallet bound to your API key, or the wallet that paid via x402. Other agents return FORBIDDEN_NOT_OWNER. Use get_a
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10日
注册名称已通过域名 cybercentry.co.uk 验证(注册中心会核实域名所有权)
代码最后更新于3个月前
我们还没有从 GitHub 读取这个仓库
在 131 毫秒内响应(HTTP 200)
无需登录
已读取 6 个工具描述,未发现针对 AI 的隐藏指令
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 4 项适用的基础检查中通过了 4 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。
你需要填写:Authorization
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3个月前,最新版本(1.0.1)于3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:Robin: AI-Powered Dark Web OSINT, HOL Guard, UniFi Network MCP
同一分类中其他作者的服务器,按星标数排序。
如何看卡片:↓ 每周下载量i★ 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
Dark web OSINT over Tor: search onion engines, scrape pages, report with your own model.
Local-first AI agent security evidence and approval workflows through HOL Guard's stdio MCP server.
Manage UniFi Network devices, clients, firewall, VLANs, VPNs, and more via MCP.
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。
Security intelligence for ERC-8004 agents: search the on-chain agent index and read verification results.
Nothing to install. This is a remote MCP server — connect to the URL and the tools appear.
| Endpoint | https://erc8126scan.ai/api/mcp |
| Transport | streamable-http |
| Registry | uk.co.cybercentry/erc-8126-scan |
| Website | https://erc8126scan.ai |
Every agent minted in the ERC-8004 Identity Registry is indexed automatically — 378,000+ and counting. Each one carries verification results across five dimensions, produced by the Cybercentry verification services:
| Dimension | What it covers | |
|---|---|---|
| ETV | Token verification | The agent's token contract |
| SCV | Solidity code verification | The staking contract source |
| MCV | Media content verification | C2PA provenance of the agent's image |
| WAV | Web application verification | The agent's live site or dApp frontend |
| WV | Wallet verification | OFAC sanctions screening and funding-chain risk |
This server exposes that index. It does not run verifications — it serves results already computed and stored.
Remote servers are added as connectors, not through claude_desktop_config.json — that file is for local servers that run a command.
https://erc8126scan.ai/api/mcpAnything that speaks MCP over streamable HTTP can connect. Point it at:
https://erc8126scan.ai/api/mcp
Check your client's own documentation for where a remote MCP server URL goes — some use a settings pane, some a config file. If yours takes a config file, the field is the server URL rather than a command.
Six tools, all read-only. None of them mints, writes on-chain, or spends your credits.
Two are free — no payment, wallet or account:
| Tool | What it returns |
|---|---|
list_services | The catalogue: every tool, its price, and how to read the risk fields |
get_index_stats | How many agents are indexed, so you can judge coverage before paying |
The other four are paid:
| Tool | What it returns | Per call | Plan |
|---|---|---|---|
search_agents | Filter the index by chain, risk score range and verification status; paginated | $0.001 | Starter |
get_agent | Full detail for one agent by UUID | $0.001 | Starter |
get_agent_activity | On-chain activity and liveness for one agent | $0.002 | Starter |
get_agent_report | Full security assessment: per-dimension findings and evidence | $0.05 | Pro |
Call list_services first. It is free, and authoritative about current pricing.
get_agent_report returns data only for agents owned by the wallet bound to your API key. Other agents return FORBIDDEN_NOT_OWNER. Public scores for any agent are available through get_agent.
Tools you cannot reach on your plan are hidden from tools/list rather than listed and refused.
Judge an agent by `risk_level`, not `overall_risk_score`.
Scoring presumes risk: a check that applies to an agent but has not been verified counts as 100. So an agent nobody has paid to assess scores exactly the same as one assessed and found dangerous. Most of the index is the former.
risk_level separates them:
| Value | Meaning |
|---|---|
not_verified | Nobody has assessed this agent yet. Not a finding against it. |
not_assessed | No score recorded |
low / moderate / elevated / high | Verified, scoring in that band |
critical | Verified and genuinely high risk |
Treating a not_verified agent as dangerous is the most common way to misread this data.
Two free tools, then two ways to pay for the rest. tools/list is public, so an agent can see everything on offer and what it costs before deciding to pay for anything.
x402 — no account, no key. The client is quoted a price, pays in USDC, and retries. Pay on whichever chain you hold funds on:
Base eip155:8453 USDC
Solana mainnet USDC
| Tool | Per call |
|---|---|
list_services | free |
get_index_stats | free |
search_agents | $0.001 |
get_agent | $0.001 |
get_agent_activity | $0.002 |
get_agent_report | $0.05 |
An unpaid call returns a JSON-RPC error with code 402, carrying the payment requirements in error.data — the same shape as the Cybercentry verification server.
API key — for volume, or clients that do not speak x402 yet. Pass an erc8126_ key as a bearer token:
Authorization: Bearer erc8126_...
Keys come from erc8126scan.ai/my-api-keys. Calls draw on your plan's daily quota instead of being paid for individually, and tool availability and page sizes follow your tier. A key that is presented but invalid is rejected outright rather than quietly falling back to per-call billing.
Either way, get_agent_report is scoped to agents owned by the caller — the wallet bound to your key, or the wallet that paid.
Cybercentry Limited