Verifa

io.github.Verifa-Solutions由 GitHub 账号 Verifa-Solutions 发布v0.1.1更新于 Oct 24, 2026

Identity verification, document checks, sanctions/PEP screening and case review as agent tools.

已验证Streamable HTTP可网页运行安全与监控商业与电商

概览

AI 生成的概览

Verifa 将身份验证、证件核查、制裁与 PEP 筛查以及案件复核作为工具提供给助手调用。

可以做什么
创建、读取、列出和重新处理验证会话,包括沙箱模拟和给申请人的采集链接,并查看会话事件与制裁、PEP、负面媒体、监控名单等核查结果及重跑。搜索、列出、获取身份并添加或移除标签。处理复核队列:列出、获取、认领、指派、评论、批准、拒绝或升级案件,以及确认或忽略发现项。读取和写入黑名单与自定义名单,列出并触发工作流和验证策略。查看组织信息,如用量统计、用户和 API 密钥,并提供连接器式的搜索与抓取。
适用场景
适合让助手端到端推进 KYC 或身份验证流程:发起会话、对人员做制裁与 PEP 筛查,并协助复核人员处理案件和发现项。建议先用沙箱密钥了解完整流程,再接触真实数据。
运行要求
需要 VERIFA_API_KEY 中的 Verifa API 密钥(沙箱或正式,可发布密钥会被拒绝),可选设置 VERIFA_ENV 为 sandbox 或 live。可作为本地 npm 包通过 stdio 运行,需 Node 20+;也可直接连接托管的 streamable HTTP 端点并使用 OAuth 2.1 登录。需要能访问 Verifa API 主机。
安装前请注意
密钥会作为 Bearer 令牌发送到 API 主机,并拥有其全部权限范围;评估时使用沙箱密钥,正式密钥应尽量缩小范围。工具可写入:会话、案件决定、评论、名单变更和工作流触发。脱敏工具默认关闭,需要 redact:write 权限和书面理由,有频率限制,且不可撤销。每次调用都会以代理身份记入审计。

安装

在 SourceWeft 中

  1. 打开 控制台中的 Verifa,将其添加到工作区。
  2. 为需要使用其工具的对话启用该服务。

Web executable,通过 Streamable HTTP。 远程服务在工作区中配置后即可从网页运行时运行。

交给智能体安装

把这段提示词粘贴到 Claude Code、Codex、Cursor 等智能体,或 SourceWeft 对话中。智能体会读取该服务器的安装指南,先向你展示发布者、端点和登录方式,经你同意后再把它添加到你的客户端。密钥不会写进任何命令。

阅读 https://sourceweft.com/mcp/io.github.Verifa-Solutions%2Fverifa/install.md,并按其中说明把这个 MCP 服务器接入我的客户端。修改任何配置之前,先告诉我它的发布者、端点和登录方式,等我确认。不要把 API 密钥写进命令或我会提交的文件;问我密钥应从哪里获取。

查看智能体所依据的安装指南

其他 MCP 客户端

把它添加到你客户端的 mcpServers 配置中。

{
  "mcpServers": {
    "verifa": {
      "type": "http",
      "url": "https://api.withverifa.com/mcp"
    }
  }
}

README

Verifa MCP Server

Run identity verification from any MCP client — Claude Desktop, Cursor, Claude Code, or your own agent. Create verification sessions, hand the person a capture link, screen against sanctions and PEP lists, read results, and work the review queue, all as tools an agent can call.

json
{  "mcpServers": {    "verifa": {      "command": "npx",      "args": ["-y", "@verifasolutionsinc/mcp-server"],      "env": { "VERIFA_API_KEY": "vk_sandbox_...", "VERIFA_ENV": "sandbox" }    }  }}

That is the whole install. Get a sandbox key at app.withverifa.com → Developers → API keys (free, no card, no sales call). Sandbox keys never touch real data and can simulate every verification outcome, so you can see the full lifecycle in a few minutes — examples/id-check does exactly that.

How it works

This package is deliberately small. It opens one authenticated connection to Verifa's hosted MCP endpoint (https://api.withverifa.com/mcp, streamable HTTP) and forwards tools/list, tools/call, prompts/list and prompts/get over stdio. Nothing is reimplemented here: tool schemas, PII scrubbing, rate limits and the audit trail all live server-side, so this wrapper never drifts from the product.

If your client already speaks streamable HTTP with OAuth 2.1, you can skip this package and connect to the hosted endpoint directly — same tools, sign-in instead of an API key.

Tools

The server exposes 45 tools, grouped into toolsets. By default you get everything except destructive.

ToolsetWhat an agent can do
sessionscreate_session, get_session, list_sessions, list_session_events, simulate_session (sandbox), reprocess_session
checkslist_checks, get_check, list_check_hits, rerun_check — sanctions, PEP, adverse media, watchlists
identitiessearch_identities, list_identities, get_identity, tag / untag
caseslist_cases, get_case, claim_case, assign_case, add_case_comment, approve_case, reject_case, escalate_case …
findingslist_findings, get_finding, acknowledge_finding, dismiss_finding
listsblocklist and custom list reads and writes
workflowslist_workflows, get_workflow, trigger_workflow, list_verification_policies
orgwhoami, get_usage_stats, list_org_users, list_api_keys
searchsearch / fetch for ChatGPT-style connectors
destructiveredact_session, redact_identity, bulk_redact_sessions, … — off unless you opt in

Every tool is scoped: the API key's scopes cap what the connection can do, whatever toolsets are enabled. Every call — read or write — writes an audit row that records it was made by an agent and by which key or OAuth client, so an agent's decision is never mistaken for a human's.

Narrowing the surface

Fewer tools means less context spent per request and better tool selection. Pick what the job needs:

json
"env": {  "VERIFA_API_KEY": "vk_sandbox_...",  "VERIFA_MCP_TOOLSETS": "sessions,checks",  "VERIFA_MCP_READ_ONLY": "1"}
VariableEffect
VERIFA_API_KEYRequired. vk_sandbox_… or vk_live_…. Publishable keys (vk_pub_…) are rejected.
VERIFA_ENVsandbox or live. Inferred from the key if omitted; must match it if set.
VERIFA_MCP_TOOLSETSComma-separated toolsets to expose. Default: all except destructive.
VERIFA_MCP_READ_ONLY1 hides every tool that writes.
VERIFA_MCP_ALLOW_DESTRUCTIVE1 adds the redaction tools. Each call also needs the redact:write scope, a written reason, and is capped at 5 per hour per key. Redaction is irreversible by design.
VERIFA_API_URLOverride the API host (e.g. a staging environment).

Check what a given configuration exposes without opening a client:

sh
VERIFA_API_KEY=vk_sandbox_... npx -y @verifasolutionsinc/mcp-server --check

Try it in five minutes

sh
git clone https://github.com/Verifa-Solutions/verifa-mcp-servercd verifa-mcp-server/examples/id-checknpm installcp .env.example .env        # paste your sandbox keynpm start                   # approvenpm run review              # the outcome that matters: a human has to decide

The demo creates a session, generates the one-time capture link and QR code an integrator would show, simulates an outcome (no passport needed — sandbox only), and reads the result. It prints the REST calls in one column and the identical flow driven through this MCP server in the other. Add --webhooks with a public URL (cloudflared tunnel --url http://localhost:8787 works) to receive the signed webhook deliveries and see signature verification — the part most integrations get wrong.

Development

sh
npm installnpm run buildnpm test                                     # unit testsVERIFA_API_KEY=vk_sandbox_... node dist/index.js --check   # live smoke test

Node 20+. Logs go to stderr; stdout is the MCP transport.

Security

  • Use sandbox keys while evaluating. Live keys should be scoped to the toolsets the agent needs.
  • The server never stores or logs the API key; it is sent only as a Bearer token to VERIFA_API_URL.
  • Responses are PII-free by default — the hosted server strips applicant identifiers from tool results unless the key holds the relevant scope and the session's sensitive-data window is open.
  • Report vulnerabilities via withverifa.com/security/disclosure.

License

MIT — see LICENSE.

来源:README.md,提交 1d9f20f

工具

0
SourceWeft 尚未读取此服务器的工具。

版本历史

1
  1. v0.1.1最新Sep 16, 2026

评分

已满足适用检查项的 85%

该评分衡量此收录条目的完整程度,依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价,也不表示服务器安全或当前可用:可连通性是带日期的检查结果,评分也不包含安装量或用户评价。

  • 近 30 天内可连通必需25 分
  • 已声明认证方式必需10 分
  • 已列出工具必需15 分

    请让服务器在无凭据时响应 tools/list,或发布可在沙箱中无凭据启动的软件包。

  • 描述必需15 分
  • 图标10 分
  • 文档15 分
  • 许可证10 分