{
  "markdown": "# Global Database — MCP Server\n\n[![Add to Cursor](https://cursor.com/deeplink/mcp-install-dark.svg)](cursor://anysphere.cursor-deeplink/mcp/install?name=global-database&config=eyJ0eXBlIjogImh0dHAiLCAidXJsIjogImh0dHBzOi8vbWNwLmdsb2JhbGRhdGFiYXNlLmNvbS9tY3AifQ==)\n[![Install in VS Code](https://img.shields.io/badge/VS_Code-Install_server-0098FF?logo=visualstudiocode&logoColor=white)](https://insiders.vscode.dev/redirect/mcp/install?name=global-database&config=%7B%22type%22%3A%20%22http%22%2C%22url%22%3A%20%22https%3A%2F%2Fmcp.globaldatabase.com%2Fmcp%22%7D)\n[![Smithery](https://img.shields.io/badge/Smithery-global--database-7c3aed)](https://smithery.ai/servers/global-database)\n[![MCP Registry](https://img.shields.io/badge/MCP_Registry-com.globaldatabase%2Fmcp-1f6feb)](https://registry.modelcontextprotocol.io)\n\nRemote [MCP](https://modelcontextprotocol.io/) server that gives LLM agents access to\n[Global Database](https://globaldatabase.com/): company profiles, financials, ownership,\ndigital insights, people search and contact enrichment, prospecting, and KYB/compliance\nlookups against official government registries.\n\nHosted at **`https://mcp.globaldatabase.com/mcp`** (Streamable HTTP). Nothing to install\nor run — you connect to it and sign in with your Global Database API key through the\nbrowser.\n\n## Install\n\n### Cursor\n\nAdd to `~/.cursor/mcp.json` (global) or `.cursor/mcp.json` (per project):\n\n```json\n{\n  \"mcpServers\": {\n    \"global-database\": {\n      \"type\": \"http\",\n      \"url\": \"https://mcp.globaldatabase.com/mcp\"\n    }\n  }\n}\n```\n\n### Claude Code\n\n```bash\nclaude mcp add --transport http global-database https://mcp.globaldatabase.com/mcp\n```\n\nOr install the plugin, which adds the server **and** the `company-due-diligence` skill:\n\n```\n/plugin marketplace add global-database/mcp-server\n/plugin install global-database@global-database\n```\n\n### VS Code (GitHub Copilot)\n\nUse the **Install server** badge above, or add it from the terminal:\n\n```bash\ncode --add-mcp '{\"name\":\"global-database\",\"type\":\"http\",\"url\":\"https://mcp.globaldatabase.com/mcp\"}'\n```\n\n### Gemini CLI\n\nInstall the bundled extension straight from this repo:\n\n```bash\ngemini extensions install https://github.com/global-database/mcp-server\n```\n\nOr add the server to `settings.json` by hand:\n\n```json\n{\n  \"mcpServers\": {\n    \"global-database\": {\n      \"httpUrl\": \"https://mcp.globaldatabase.com/mcp\",\n      \"authProviderType\": \"dynamic_discovery\"\n    }\n  }\n}\n```\n\n### Perplexity\n\nRequires a paid plan (custom connectors are a Pro/Enterprise feature). In **Settings →\nConnectors → + Custom connector → Remote**, enter:\n\n| Field | Value |\n|---|---|\n| Name | Global Database |\n| MCP Server URL | `https://mcp.globaldatabase.com/mcp` |\n| Authentication | OAuth |\n| Transport | Streamable HTTP |\n\nThen open the connector card to run the OAuth sign-in.\n\n### Grok\n\nAt [grok.com/connectors](https://grok.com/connectors) → **New Connector → Custom**, enter the\nMCP server URL `https://mcp.globaldatabase.com/mcp` and complete the OAuth sign-in. Grok\ndiscovers the tools from the live endpoint.\n\n### Microsoft Copilot Studio\n\nIn your agent → **Tools → Add a tool → Model Context Protocol → New MCP server**, enter the\nserver URL `https://mcp.globaldatabase.com/mcp` and pick **OAuth 2.0** with dynamic\nregistration / discovery. Leave **Scope empty** — the server publishes no scopes, and a\nnon-empty one fails the token request.\n\nCopilot Studio ignores the MCP `instructions` field, so the agent starts without the routing\nrules every other client receives. Paste\n[`copilot-studio-instructions.md`](copilot-studio-instructions.md) into **Agent → Overview →\nInstructions** after the connection is created.\n\n### Claude.ai / other MCP clients\n\nAdd a custom connector pointing at `https://mcp.globaldatabase.com/mcp`.\n\n## Authentication\n\nOAuth 2.1 with PKCE and Dynamic Client Registration — the client registers itself, so there\nis nothing to configure. On first connect your browser opens a Global Database login page\nwhere you paste your API key; the key is held in the access-token claims server-side and is\nnever passed as a tool argument.\n\nGet an API key at [globaldatabase.com](https://globaldatabase.com/).\n\n## Tools\n\n| Tool | What it does |\n|---|---|\n| `check_api_key` | Verify the API connection and return current user info. |\n| `get_company_by_website` | Look up a company by website URL or domain. |\n| `get_company_by_linkedin` | Look up a company by LinkedIn URL or public ID. |\n| `get_company_by_identifiers` | Look up a company by name, registration number, VAT, ticker, website, email, or LinkedIn. |\n| `get_company_details` | Full company profile by `company_id`. |\n| `get_company_financials` | Balance sheet, ratios, and key metrics per year. |\n| `get_company_ownership` | Shareholders and corporate group structure. |\n| `get_digital_insights` | Web traffic, rankings, traffic sources, WHOIS, technologies. |\n| `search_employees` | Find the people at a company, or people by name, role, department, or seniority. |\n| `get_employee_details` | Full profile for one person from a `search_employees` row. |\n| `find_people_by_domain` | Find people at companies given by web domain, via the contact-enrichment providers. |\n| `enrich_employee_contacts` | Find and enrich one named employee/contact — email, phone, socials. |\n| `get_nomenclature` | Lookup values for use with prospecting filters. |\n| `prospecting` | Search and filter companies by country, industry, size, revenue, and more. |\n| `kyb_search` | Search official government registries for KYB/compliance checks. |\n| `kyb_company_details` | Official registry details for a company. |\n| `kyb_officers` | Directors, secretaries, and officers from the official registry. |\n| `kyb_shareholders` | Shareholders from the official registry. |\n| `kyb_shareholders_search` | Reverse shareholder lookup by holder name, across jurisdictions. |\n| `kyb_group_structure` | Corporate group structure from the official registry. |\n| `kyb_financial` | Detailed financial statements across multiple years. |\n| `kyb_officers_search` | Reverse officer lookup by person's name, across jurisdictions. |\n\n## Resources and prompts\n\nBeyond tools, the server exposes MCP resources so a client can load reference data once per\nconversation instead of spending a tool call (and credits) on it:\n\n| Resource | Contents |\n|---|---|\n| `gdb://nomenclature/country` | Country list for the `prospecting` location filter. |\n| `gdb://nomenclature/kyb-countries` | Country list for `kyb_search`. |\n| `gdb://nomenclature/company-status` | Company status values for `prospecting`. |\n| `gdb://nomenclature/department` | Department ids for `search_employees`. |\n| `gdb://nomenclature/seniority` | Seniority ids for `search_employees`. |\n| `gdb://docs/prospecting-filters` | Full prospecting filter catalog — field names and value formats. |\n| `gdb://docs/intent-glossary` | Keyword → tool glossary for routing a request. |\n\nLarger, hierarchical nomenclatures (NACE, ISIC, SIC, industry focus, region) stay behind the\n`get_nomenclature` tool and its `search` parameter.\n\nPrompts shipped with the server: `company_research`, `due_diligence`, `lead_generation`,\n`competitor_analysis`, `market_research`.\n\n## Plugins and skill\n\nThe same contents ship as a plugin for both Cursor (`.cursor-plugin/plugin.json`) and\nClaude Code (`.claude-plugin/marketplace.json`), so the server can be installed from a\nmarketplace instead of being wired up by hand. Each plugin bundles:\n\n- the remote MCP server above (all tools), and\n- a **`company-due-diligence`** skill that walks the agent through a structured KYB\n  review — resolve the entity, then pull registry details, officers, shareholders,\n  group structure and financials into one dossier.\n\n## Where it is listed\n\nVerified 2026-09-01.\n\n| Directory | Entry |\n|---|---|\n| [Official MCP Registry](https://registry.modelcontextprotocol.io) | [`com.globaldatabase/mcp`](https://registry.modelcontextprotocol.io/v0/servers?search=globaldatabase) — published from [`server.json`](server.json) |\n| [Claude Connectors Directory](https://claude.ai/directory) | [`global-database`](https://claude.ai/directory/global-database) — Community tier |\n| [OpenAI ChatGPT](https://chatgpt.com/) | [Global Database](https://chatgpt.com/plugins/plugin_asdk_app_696f807d21a481918a1ed1f43d719ce9) — Apps SDK app |\n| [Smithery](https://smithery.ai/servers/global-database) | [`global-database`](https://smithery.ai/servers/global-database) |\n| [Glama](https://glama.ai/mcp/servers) | [`com.globaldatabase/mcp`](https://glama.ai/mcp/connectors/com.globaldatabase/mcp) |\n| [Cursor Directory](https://cursor.directory/) | [`mcp-global-database-3`](https://cursor.directory/plugins/mcp-global-database-3) |\n| [mcptop.com](https://mcptop.com/) | [`openai-global-database`](https://mcptop.com/server/openai-global-database) — third-party leaderboard |\n\nAlso installable without a directory:\n\n| Channel | How |\n|---|---|\n| Claude Code plugins | `/plugin marketplace add global-database/mcp-server` |\n| Antigravity CLI | `agy plugin install` against this repo |\n| Gemini Enterprise | admin registers it as a custom MCP connection per tenant |\n\nNot listed: PulseMCP (their submissions are paused), mcp.so, and the VS Code / GitHub MCP\ngallery — that last one is a curated catalogue of a few hundred servers and is **not** an\ningest of the Official MCP Registry, so publishing there does not reach it.\n\nMaintainer notes — publishing, DNS verification and the remaining directories:\n[`PUBLISHING.md`](PUBLISHING.md).\n\n## Links\n\n- [Homepage](https://mcp.globaldatabase.com)\n- [Smithery listing](https://smithery.ai/servers/global-database)\n- [Privacy policy](https://mcp.globaldatabase.com/static/privacy.html)\n- [Terms](https://mcp.globaldatabase.com/static/terms.html)\n\n## About this repository\n\nDistribution metadata only (`server.json`, `.mcp.json`, `plugin.json`,\n`.claude-plugin/`, `.cursor-plugin/`, `gemini-extension.json`) so MCP directories and\nclients can discover the hosted server, plus the client-side extras that ship with it — the\n`company-due-diligence` skill and the Copilot Studio instruction block. MIT covers this\nmetadata; the server implementation and the Global Database API are not open source.\n",
  "bytes": 10234,
  "sha": "208d32ff88de0c28252fb8063b7db17fd9c5f8c555c9c4a61e6520011ba7d623",
  "repo_slug": "global-database/mcp-server",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_com_globaldatabase_mcp_f1f4a9eb/readme"
}