{
  "markdown": "<div align=\"center\">\n\n<h1>Open Enthrium AI MCP Server</h1>\n<h3>Enterprise MCP Server · Apache-2.0 · Claude Code · Cursor · Windsurf · Codex · Claude Desktop · VS Code</h3>\n\n**Connect any AI coding assistant to your enterprise data — databases, files, APIs, and more — via a single binary.**\n\n[![License: Apache 2.0](https://img.shields.io/badge/License-Apache%202.0-4f46e5.svg)](LICENSE)\n[![GitHub Release](https://img.shields.io/github/v/release/enthrium/open-enthrium-ai-mcp-server?color=4f46e5&label=latest)](https://github.com/enthrium/open-enthrium-ai-mcp-server/releases)\n[![Windows](https://img.shields.io/badge/Windows-Download-0078D4?logo=windows&logoColor=white)](https://github.com/enthrium/open-enthrium-ai-mcp-server/releases/latest/download/oe-mcp-win.exe)\n[![Linux](https://img.shields.io/badge/Linux-Download-E95420?logo=linux&logoColor=white)](https://github.com/enthrium/open-enthrium-ai-mcp-server/releases/latest/download/oe-mcp-linux)\n[![macOS](https://img.shields.io/badge/macOS-Download-000000?logo=apple&logoColor=white)](https://github.com/enthrium/open-enthrium-ai-mcp-server/releases/latest/download/oe-mcp-macos)\n[![npm](https://img.shields.io/npm/v/@openenthrium/oe-mcp?color=0284c7&label=npm)](https://www.npmjs.com/package/@openenthrium/oe-mcp)\n[![Website](https://img.shields.io/badge/Website-openenthrium.com-4f46e5)](https://www.openenthrium.com)\n\n</div>\n\n---\n\n## What is OE MCP Server?\n\nA standalone binary that implements the [Model Context Protocol (MCP)](https://modelcontextprotocol.io) and exposes your enterprise data sources as tools that AI apps can use directly. No code. Define connectors in a single JSON file.\n\n- **45+ connector categories** — PostgreSQL, MongoDB, S3, GitHub, Slack, Gmail, SSH, REST API, and more\n- **Two transport modes** — `--stdio` for Claude Code / Cursor / Windsurf; `--serve` for cloud or team deployments\n- **Persistent memory** — `memory_set / memory_get / memory_list / memory_delete` survive across sessions\n- **Action log** — every connector call logged automatically with timestamp, tool, input, and result\n- **Run AI agents** — `run_agent` executes any OE Runtime SKILL.md agent directly from Claude Code, Cursor, or any MCP client. Manual skills pause for approval via `approve_chain`.\n- **Self-hosted** — runs on your own machine, no cloud dependency, no call-home\n\n---\n\n## Quick Start via npm (Recommended)\n\n**1. Create `oe-mcp.json`:**\n\n```json\n{\n  \"connectors\": [\n    {\n      \"name\": \"my-postgres\",\n      \"type\": \"postgresql\",\n      \"host\": \"localhost\",\n      \"port\": 5432,\n      \"database\": \"mydb\",\n      \"user\": \"postgres\",\n      \"password\": \"secret\"\n    },\n    {\n      \"name\": \"my-codebase\",\n      \"type\": \"filesystem\",\n      \"basePath\": \"/home/user/projects/myapp\"\n    }\n  ],\n  \"memory\": [\n    { \"key\": \"project_context\", \"value\": \"This is our main application.\" }\n  ]\n}\n```\n\n**2. Add to your AI app's MCP config:**\n\nmacOS / Linux:\n```json\n{\n  \"mcpServers\": {\n    \"oe-mcp\": {\n      \"type\": \"stdio\",\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@openenthrium/oe-mcp\", \"--stdio\", \"/path/to/oe-mcp.json\"]\n    }\n  }\n}\n```\n\nWindows:\n```json\n{\n  \"mcpServers\": {\n    \"oe-mcp\": {\n      \"type\": \"stdio\",\n      \"command\": \"npx.cmd\",\n      \"args\": [\"-y\", \"@openenthrium/oe-mcp\", \"--stdio\", \"C:\\\\path\\\\to\\\\oe-mcp.json\"]\n    }\n  }\n}\n```\n\n> **`-y` is required** — without it npx blocks waiting for keyboard input and the MCP connection never opens.\n\n**3. Reload your AI app** — connectors appear as tools automatically.\n\nAsk Claude: _\"What connectors do you have access to?\"_ to verify.\n\n---\n\n## Download (Standalone Binary)\n\n| Platform | Binary |\n|---|---|\n| **Windows** | [oe-mcp-win.exe](https://github.com/enthrium/open-enthrium-ai-mcp-server/releases/latest/download/oe-mcp-win.exe) |\n| **Linux** | [oe-mcp-linux](https://github.com/enthrium/open-enthrium-ai-mcp-server/releases/latest/download/oe-mcp-linux) |\n| **macOS** | [oe-mcp-macos](https://github.com/enthrium/open-enthrium-ai-mcp-server/releases/latest/download/oe-mcp-macos) |\n| **Sample configs** | [oe-mcp-samples.zip](https://github.com/enthrium/open-enthrium-ai-mcp-server/releases/latest/download/oe-mcp-samples.zip) |\n\n---\n\n## Built-in Tools\n\n### Memory\n\nPersistent memory that survives restarts — stored in `oe-mcp-memory.json`:\n\n| Tool | Description |\n|---|---|\n| `memory_set` | Store a key-value pair across sessions |\n| `memory_get` | Retrieve a stored value by key |\n| `memory_list` | List all stored key-value pairs |\n| `memory_delete` | Remove a stored key |\n\n> _\"Remember that our production database is on prod-db.company.com\"_ → Claude calls `memory_set`\n\n### Action Log\n\nEvery connector call is logged automatically to `oe-mcp-log.json`:\n\n| Tool | Description |\n|---|---|\n| `log_list` | List recent connector calls (newest first, supports `limit`) |\n| `log_clear` | Clear all log entries |\n\n### Run AI Agents\n\nExecute OE Runtime SKILL.md agents or YAML agents directly from Claude Code, Cursor, or any MCP client — no terminal required:\n\n| Tool | Description |\n|---|---|\n| `run_agent` | Run an agent by file path. Auto skills execute immediately; manual skills pause and return `pending_skill_chain`. |\n| `list_pending_skills` | List all manual skills currently paused and waiting for approval |\n| `approve_chain` | Approve, skip, or abort a paused manual skill by `chain_id` |\n\n**`run_agent` parameters:**\n\n| Parameter | Required | Description |\n|---|---|---|\n| `file` | ✅ | Absolute path to `agent.yaml` |\n| `params` | ❌ | Key-value pairs substituted via `{{key}}` in the agent |\n| `input` | ❌ | Optional initial message passed to the agent |\n\n**`approve_chain` parameters:**\n\n| Parameter | Required | Description |\n|---|---|---|\n| `chain_id` | ✅ | From `pending_skill_chain.chain_id` in a `run_agent` response |\n| `approved` | ❌ | `true` to run the skill (default), `false` to skip it and continue |\n| `abort` | ❌ | `true` to stop the entire pipeline immediately |\n\nOE MCP looks for `oe-config.json` in the agent's directory first, then falls back to `oe-mcp.json`.\n\n### Agent Skill Approval Flow\n\nWhen an agent's skill pipeline includes manual skills, Claude handles the approval loop automatically:\n\n1. Claude calls `run_agent` → response shows `⏸ Skill awaiting approval` with `chain_id` and `skill_name`\n2. Claude decides — based on your instructions — whether to approve, skip, or abort\n3. Claude calls `approve_chain` → next skill runs or the next manual skill pauses again\n4. Repeat until `Pipeline complete` or Claude aborts\n\nExample instruction to Claude Code: _\"Run the OE Skills orchestrator and send a Slack message — skip anything you can't do, abort if it asks for credentials.\"_\n\n---\n\n## Transport Modes\n\n| Mode | Flag | Best for |\n|---|---|---|\n| **stdio** | `--stdio` | Claude Code, Cursor, Windsurf, Codex, Claude Desktop — launched as child process |\n| **HTTP** | `--serve --port 4040` | Cloud deployments, sharing one server across a team |\n\nHTTP mode — start the server, then add the URL to Cursor / Windsurf / Claude Desktop:\n```bash\noe-mcp-linux --serve --port 4040 /path/to/oe-mcp.json\n# → http://your-server.com:4040/mcp\n```\n\n---\n\n## Sample Configs\n\nDownload [oe-mcp-samples.zip](https://github.com/enthrium/open-enthrium-ai-mcp-server/releases/latest/download/oe-mcp-samples.zip) — ready-to-use `oe-mcp.json` for common connectors:\n\n`postgres` · `mysql` · `mongodb` · `github` · `slack` · `gdrive` · `ssh` · `filesystem` · `oracle` · `salesforce` · `servicenow` · `telegram` · `notion` · `confluence` · `graphql` · `zoho-mail` · `sftp` · `dropbox` · `multi-connector`\n\n---\n\n## Part of Open Enthrium\n\n| | |\n|---|---|\n| ⚡ **Agent Runtime** | [open-enthrium-ai-agent-runtime](https://github.com/enthrium/open-enthrium-ai-agent-runtime) — run SKILL.md agents as CLI or HTTP server |\n| 🖥️ **Platform** | [open-enthrium-ai-platform](https://github.com/enthrium/open-enthrium-ai-platform) — full web app with workspaces, RAG, Agent Builder |\n| 🌐 **Website** | [openenthrium.com](https://www.openenthrium.com) |\n\n---\n\n## Contributing\n\n→ See **[CONTRIBUTING.md](CONTRIBUTING.md)** for how to add sample configs and connector adapters.\n\n---\n\n## License\n\n[Apache-2.0](LICENSE) — free to use, modify, and deploy for any purpose, including commercial use.\nNo usage limits. No telemetry. No call-home.\n\n---\n\n<div align=\"center\">\n\n**[⭐ Star this repo](https://github.com/enthrium/open-enthrium-ai-mcp-server)** &nbsp;·&nbsp; **[🌐 Website](https://www.openenthrium.com)** &nbsp;·&nbsp; **[⚡ Agent Runtime](https://github.com/enthrium/open-enthrium-ai-agent-runtime)**\n\n</div>\n",
  "bytes": 8548,
  "sha": "82d3f769f0ed312bc168c90e95834d9a7a79f10d726a93ecf43434dca3c73f8c",
  "repo_slug": "openenterprise-info/open-enterprise-ai-mcp-server",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_openenterprise_info_oe_mcp_6643af5c/readme"
}