{
  "markdown": "---\ntitle: Arpeio MCP\nemoji: 🛠️\ncolorFrom: blue\ncolorTo: indigo\nsdk: docker\npinned: false\n---\n\n# arpeio-mcp\n\n<!-- mcp-name: io.github.arpe-io/arpeio-mcp -->\n\n[![PyPI version](https://img.shields.io/pypi/v/arpeio-mcp)](https://pypi.org/project/arpeio-mcp/)\n[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)\n\nUnified MCP server for [Arpe.io](https://arpe.io) data tools — build, preview, and execute high-performance data commands through AI assistants.\n\n| Tool | Description |\n|------|-------------|\n| **FastBCP** | High-performance parallel database export to files and cloud |\n| **FastTransfer** | High-performance parallel data transfer between databases |\n| **LakeXpress** | Automated database-to-cloud data pipeline as Parquet |\n| **MigratorXpress** | Cross-platform database migration with parallel transfer |\n\n> **No binaries required.** All tools work in **command builder mode** out of the box — command building, preview, and informational tools work without any Arpe.io binary installed. To enable execution, download the binaries from [arpe.io](https://arpe.io) and set the corresponding `*_PATH` environment variables.\n\n## Connect your AI assistant\n\nA hosted instance is available at `https://arpe-io-arpeio-mcp.hf.space/sse` — no installation required. For local installation with execution support, use the stdio transport via `pip install arpeio-mcp`.\n\n[ChatGPT](#chatgpt) | [Claude Code](#claude-code) | [Claude Desktop](#claude-desktop) | [Cursor](#cursor) | [Gemini CLI](#gemini-cli) | [HuggingChat](#huggingchat) | [Kiro IDE](#kiro-ide) | [Le Chat (Mistral)](#le-chat-mistral) | [VS Code](#vs-code-github-copilot) | [Windsurf](#windsurf)\n\n### ChatGPT\n\n*Available for paid plans only (Plus, Pro, Team, and Enterprise).*\n\n1. Open ChatGPT in your browser, go to **Settings** > **Apps and connectors**.\n2. Open **Advanced settings** and enable **Developer mode**.\n3. Go to **Connectors** > **Browse connectors** > **Add a new connector**.\n4. Set the URL to `https://arpe-io-arpeio-mcp.hf.space/sse` and save.\n\n### Claude Code\n\n```shell\nclaude mcp add --transport sse arpeio https://arpe-io-arpeio-mcp.hf.space/sse\n```\n\nOr for local installation with execution support:\n\n```shell\npip install arpeio-mcp\nclaude mcp add arpeio arpeio-mcp\n```\n\n### Claude Desktop\n\nAdd the following to your Claude Desktop configuration file:\n- **Linux**: `~/.config/Claude/claude_desktop_config.json`\n- **macOS**: `~/Library/Application Support/Claude/claude_desktop_config.json`\n- **Windows**: `%APPDATA%\\Claude\\claude_desktop_config.json`\n\n#### Remote (no installation)\n\n```json\n{\n  \"mcpServers\": {\n    \"arpeio\": {\n      \"command\": \"npx\",\n      \"args\": [\n        \"mcp-remote\",\n        \"https://arpe-io-arpeio-mcp.hf.space/sse\"\n      ]\n    }\n  }\n}\n```\n\n#### Local (with execution support)\n\n```json\n{\n  \"mcpServers\": {\n    \"arpeio\": {\n      \"command\": \"arpeio-mcp\",\n      \"env\": {\n        \"FASTBCP_PATH\": \"/path/to/FastBCP\",\n        \"FASTTRANSFER_PATH\": \"/path/to/FastTransfer\",\n        \"LAKEXPRESS_PATH\": \"/path/to/LakeXpress\",\n        \"MIGRATORXPRESS_PATH\": \"/path/to/MigratorXpress\"\n      }\n    }\n  }\n}\n```\n\n### Cursor\n\n1. Open Cursor Settings and search for \"MCP\".\n2. Add a new MCP server with the following configuration:\n\n```json\n{\n  \"mcpServers\": {\n    \"arpeio\": {\n      \"url\": \"https://arpe-io-arpeio-mcp.hf.space/sse\"\n    }\n  }\n}\n```\n\n### Gemini CLI\n\nAdd the following to your `~/.gemini/settings.json` file:\n\n```json\n{\n  \"mcpServers\": {\n    \"arpeio\": {\n      \"uri\": \"https://arpe-io-arpeio-mcp.hf.space/sse\"\n    }\n  }\n}\n```\n\n### HuggingChat\n\n1. In the chat interface, click the **+** icon, select **MCP Servers**, then **Manage MCP Servers**.\n2. Click **Add Server**.\n3. Set the **Server Name** to `Arpe.io` and the **Server URL** to `https://arpe-io-arpeio-mcp.hf.space/sse`.\n4. Click **Add Server** and verify the health check shows **Connected**.\n\n### Kiro IDE\n\nAdd the following to your Kiro MCP configuration file (`.kiro/settings/mcp.json` in your workspace):\n\n```json\n{\n  \"mcpServers\": {\n    \"arpeio\": {\n      \"url\": \"https://arpe-io-arpeio-mcp.hf.space/sse\"\n    }\n  }\n}\n```\n\n### Le Chat (Mistral)\n\n*Available on all plans, including free.*\n\n1. Go to **Intelligence** > **Connectors**.\n2. Click **Add connector** > **Custom MCP Connector**.\n3. Set the name to `Arpe.io` and the URL to `https://arpe-io-arpeio-mcp.hf.space/sse`.\n4. Leave authentication disabled and click **Create**.\n\n### VS Code (GitHub Copilot)\n\nAdd the following to your VS Code MCP configuration. Run **MCP: Open User Configuration** from the Command Palette to open it.\n- **Linux**: `~/.config/Code/User/mcp.json`\n- **macOS**: `~/Library/Application Support/Code/User/mcp.json`\n- **Windows**: `%APPDATA%\\Code\\User\\mcp.json`\n\n```json\n{\n  \"servers\": {\n    \"arpeio\": {\n      \"url\": \"https://arpe-io-arpeio-mcp.hf.space/sse\",\n      \"type\": \"sse\"\n    }\n  }\n}\n```\n\n### Windsurf\n\nAdd the following to your Windsurf configuration file:\n- **Linux**: `~/.codeium/windsurf/mcp_config.json`\n- **macOS**: `~/.codeium/windsurf/mcp_config.json`\n- **Windows**: `%USERPROFILE%\\.codeium\\windsurf\\mcp_config.json`\n\n```json\n{\n  \"mcpServers\": {\n    \"arpeio\": {\n      \"command\": \"npx\",\n      \"args\": [\n        \"-y\",\n        \"mcp-remote\",\n        \"https://arpe-io-arpeio-mcp.hf.space/sse\"\n      ]\n    }\n  }\n}\n```\n\n## Local Installation\n\nFor full execution support (not just command building), install locally:\n\n```bash\npip install arpeio-mcp\n```\n\nThen configure your AI assistant to use the `arpeio-mcp` command (stdio transport) with optional binary paths — see the [Claude Desktop local configuration](#local-with-execution-support) for an example.\n\n## Environment Variables\n\n| Variable | Description | Required |\n|----------|-------------|----------|\n| `FASTBCP_PATH` | Path to FastBCP binary | No |\n| `FASTTRANSFER_PATH` | Path to FastTransfer binary | No |\n| `LAKEXPRESS_PATH` | Path to LakeXpress binary | No |\n| `MIGRATORXPRESS_PATH` | Path to MigratorXpress binary | No |\n| `FASTBCP_DIR_PATH` | FastBCP directory for LakeXpress | No |\n| `FASTTRANSFER_DIR_PATH` | FastTransfer directory for MigratorXpress | No |\n| `*_TIMEOUT` | Per-tool execution timeout (seconds) | No |\n| `*_LOG_DIR` | Per-tool log directory | No |\n| `LOG_LEVEL` | Logging level (DEBUG/INFO/WARNING/ERROR) | No |\n\n## Available Tools (17)\n\nRead-only advisory tools were consolidated into one `*_info` tool per product (with an `action` enum) and per-product release-notes tools were merged into a single `arpe_release_notes`. Auto-parallelism is now suggested inside `preview` so the typical workflow is **2 calls** (`preview` → `execute`) instead of 4.\n\n### FastBCP (3 tools)\n- `fastbcp_info` — Read-only advisory: `action=\"formats\" | \"parallelism\" | \"workflow\" | \"version\"`\n- `fastbcp_preview_export` — Validate parameters and render the command (auto-suggests parallelism when `method` is omitted)\n- `fastbcp_execute_export` — Run the export\n\n### FastTransfer (3 tools)\n- `fasttransfer_info` — Read-only advisory: `action=\"combinations\" | \"parallelism\" | \"workflow\" | \"version\"`\n- `fasttransfer_preview_transfer` — Validate parameters and render the command (auto-suggests parallelism)\n- `fasttransfer_execute_transfer` — Run the transfer\n\n### LakeXpress (3 tools)\n- `lakexpress_info` — Read-only advisory: `action=\"capabilities\" | \"workflow\" | \"version\"`\n- `lakexpress_preview_command` — Build any LakeXpress command (`lxdb_*`, `config_*`, `sync`, `sync[export]`, `sync[publish]`, `run`, `status`, `cleanup`). On v0.4.0+ binaries, warns when `-a` / `--lxdb_auth_id` / `--sync_id` are missing on sync-family calls\n- `lakexpress_execute_command` — Run the command\n\n### MigratorXpress (4 tools)\n- `migratorxpress_info` — Read-only advisory: `action=\"capabilities\" | \"workflow\" | \"version\"`\n- `migratorxpress_validate_auth_file` — Validate the JSON auth file (only file-I/O advisory tool kept separate)\n- `migratorxpress_preview_command` — Build the migrate command. Accepts the new `project` tag (v0.6.30+); warns on `migration_db_type=\"postgres\"` against pre-0.6.32 binaries\n- `migratorxpress_execute_command` — Run the migration\n\n### Meta (4 tools)\n- `arpe_get_status` — Status of all four CLIs (installed / command-builder-only)\n- `arpe_quick_start` — Detect the right tool from a plain-English use case and return a workflow guide\n- `arpe_release_notes` — Return release-notes chunks for any product (`product=\"fastbcp\" | \"fasttransfer\" | \"lakexpress\" | \"migratorxpress\"`, optional `version`)\n- `search_docs` — BM25 full-text search over arpe.io docs sites and blog\n\n## Structured output\n\nThe command-building, execution, and discovery tools return **structured content**\n(`outputSchema` + `structuredContent`) alongside the human-readable markdown, so\nMCP clients can chain calls programmatically instead of parsing prose:\n\n- **`*_preview_*`** → the built `command` (argv), `command_string`, masked `command_display`, `explanation`, and version `warnings`. The exact command can be handed straight to the matching `*_execute_*` tool.\n- **`*_execute_*`** → `success`, `return_code`, `stdout`, `stderr`, and parsed `diagnostics`.\n- **`search_docs`** → ranked `results` records; **`arpe_get_status`** → per-product status; **`arpe_release_notes`** → release-notes `chunks`.\n\nEach payload carries a `status` field (`ok` / `error`) so success and error responses are both machine-checkable.\n\n## Prompts (5)\n\nConversation starters surfaced by clients that support MCP prompts (Claude Desktop, Cursor, etc.):\n`export-table`, `transfer-data`, `lakehouse-pipeline`, `migrate-database`, `troubleshoot`.\n\n## Resources (4)\n\nStatic capability matrices served as MCP resources so clients can prefetch them without a tool call:\n`arpeio://capabilities/fastbcp-formats`, `fasttransfer-combinations`, `lakexpress-capabilities`, `migratorxpress-capabilities`.\n\n## Evaluations\n\n`evaluations/arpeio_eval.xml` holds 10 read-only, verifiable questions used to check\nthat an LLM can drive the server to correct answers (tool selection, parallelism\nrecommendations, capability lookups, version gating). A guard test re-derives every\nanswer from the capability registries so the answer key stays honest:\n\n```bash\npython -m pytest tests/test_evaluations.py -q\n```\n\n## License\n\nMIT\n",
  "bytes": 10254,
  "sha": "d2115690089c000eb4e53c117d00919c1517a284eda0a2d963ab4b7f0ae7bdfb",
  "repo_slug": "arpe-io/arpeio-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_arpe_io_arpeio_mcp_7b3b3ca7/readme"
}