{"id":50507589,"url":"https://github.com/barateza/mcp-plesk-dev-docs","last_synced_at":"2026-06-02T17:30:47.322Z","repository":{"id":337270005,"uuid":"1152924656","full_name":"barateza/mcp-plesk-dev-docs","owner":"barateza","description":"A unified MCP server that indexes and retrieves Plesk documentation using vector embeddings and semantic search with reranking.","archived":false,"fork":false,"pushed_at":"2026-05-22T06:03:32.000Z","size":1595,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-05-22T13:43:15.419Z","etag":null,"topics":["documentation","embeddings","fastmcp","mcp","model-context-protocol","plesk","plesk-extension","python","rag","semantic-search","vector-database"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/barateza.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.md","contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":"SECURITY.md","support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":"NOTICE","maintainers":null,"copyright":null,"agents":"AGENTS.md","dco":null,"cla":null}},"created_at":"2026-02-08T16:38:52.000Z","updated_at":"2026-05-22T06:03:35.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/barateza/mcp-plesk-dev-docs","commit_stats":null,"previous_names":["barateza/mcp-plesk-unified","barateza/mcp-plesk-dev-docs"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/barateza/mcp-plesk-dev-docs","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/barateza%2Fmcp-plesk-dev-docs","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/barateza%2Fmcp-plesk-dev-docs/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/barateza%2Fmcp-plesk-dev-docs/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/barateza%2Fmcp-plesk-dev-docs/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/barateza","download_url":"https://codeload.github.com/barateza/mcp-plesk-dev-docs/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/barateza%2Fmcp-plesk-dev-docs/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":33833277,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-06-02T02:00:07.132Z","response_time":109,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["documentation","embeddings","fastmcp","mcp","model-context-protocol","plesk","plesk-extension","python","rag","semantic-search","vector-database"],"created_at":"2026-06-02T17:30:46.447Z","updated_at":"2026-06-02T17:30:47.315Z","avatar_url":"https://github.com/barateza.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# mcp-plesk-dev-docs\n\n[![Python 3.12+](https://img.shields.io/badge/python-3.12%2B-blue?style=flat-square)](https://www.python.org/downloads/)\n[![PyPI Version](https://img.shields.io/pypi/v/mcp-plesk-dev-docs?style=flat-square\u0026color=blue)](https://pypi.org/project/mcp-plesk-dev-docs/)\n[![PyPI Downloads](https://img.shields.io/pypi/dm/mcp-plesk-dev-docs?style=flat-square\u0026color=blue)](https://pypi.org/project/mcp-plesk-dev-docs/)\n[![MCP Registry](https://img.shields.io/badge/MCP%20Registry-listed-green?style=flat-square)](https://registry.modelcontextprotocol.io/v0.1/servers/io.github.barateza/mcp-plesk-dev-docs)\n[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg?style=flat-square)](LICENSE)\n[![MCP Compatible](https://img.shields.io/badge/MCP-Compatible-green?style=flat-square)](https://modelcontextprotocol.io/)\n[![Code style: black](https://img.shields.io/badge/code%20style-black-000000.svg?style=flat-square)](https://github.com/psf/black)\n[![Ruff](https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/astral-sh/ruff/main/assets/badge/v2.json\u0026style=flat-square)](https://github.com/astral-sh/ruff)\n[![MCP Badge](https://lobehub.com/badge/mcp/barateza-mcp-plesk-unified)](https://lobehub.com/mcp/barateza-mcp-plesk-unified)\n\n\u003e [!NOTE]\n\u003e This MCP server provides unified documentation search for extension developers. If you are looking to manage your live Plesk server via AI, please see the [official Plesk MCP Server](https://talk.plesk.com/threads/mcp-server-extension-for-plesk-wp-toolkit.389895/).\n\n**State-of-the-Art (SOTA) semantic search across the entire Plesk documentation surface, optimized for sub-second latency on Apple Silicon.**\n\n---\n\n## Why this exists\n\nPlesk documentation is spread across five separate sources: an admin guide, a REST API reference, a CLI reference, a PHP SDK, and a JS SDK. Answering a single extension development question often means searching all of them manually, cross-referencing results, and still missing the relevant section.\n\nThis server ingests all five sources, embeds them with a multilingual model, and exposes a single `search_plesk_unified` MCP tool. It uses hybrid search (Vector + FTS), Reciprocal Rank Fusion (RRF), and Cross-Encoder reranking to deliver high-precision results in milliseconds.\n\n---\n\n## Architecture \u0026 Performance\n\n```mermaid\nflowchart TD\n    Client[\"MCP Client\\n(Claude Desktop / Cursor / etc.)\"]\n\n    Client --\u003e|\"search_plesk_unified(query)\"| Server\n\n    subgraph Server[\"FastMCP Server · Modular Architecture\"]\n        direction TB\n        Main[\"Bootstrap · server/main.py\"]\n        Life[\"Lifecycle Hooks · server/lifecycle.py\"]\n        Tools[\"MCP Tools · server/mcp_app.py\"]\n\n        Main --\u003e Life --\u003e Tools\n    end\n\n    subgraph Pipeline[\"Retrieval Pipeline\"]\n        direction TB\n        E[\"1 · Embed query\\n(Hardware-accelerated)\"]\n        S[\"2 · Hybrid Search\\nVector (LanceDB) + FTS (Tantivy)\"]\n        R[\"3 · RRF Merge + Rerank\\n(MiniLM-L4-v2)\"]\n        N[\"4 · Neighbor Expansion\\n(Context Enrichment)\"]\n        A[\"5 · AI Synthesis\\n(sampling-enabled)\"]\n        E --\u003e S --\u003e R --\u003e N --\u003e A\n    end\n\n    subgraph Store[\"LanceDB Vector \u0026 FTS Store\"]\n        direction LR\n        G[\"Guide\"]\n        A_[\"API\"]\n        C[\"CLI\"]\n        P[\"PHP Stubs\"]\n        J[\"JS SDK\"]\n    end\n\n    Tools --\u003e Pipeline\n    S \u003c--\u003e Store\n```\n\n### Performance Benchmarks (2026-05-04)\nOptimized for Apple Silicon (M2/M3) using MPS acceleration and memory-resident table caching.\n\n| Profile | Embed Model | HR@5 | MRR@5 | Avg Latency | Est. RAM |\n| :--- | :--- | :--- | :--- | :--- | :--- |\n| **`light`** | BAAI/bge-small | **100.0%** | **0.917** | **1.007 s** | ~200 MB |\n| **`medium`** | BAAI/bge-base | **100.0%** | **0.917** | **~0.60s** | ~600 MB |\n| **`full-tq`** | BAAI/bge-m3 | 75.0% | 0.750 | **~0.40s** | ~1300 MB |\n\n*Metrics measured on Apple M2 Pro with LanceDB connection caching enabled.*\n\n---\n\n## Key Features\n\n- **Single-Instance Lock:** PID-based lock prevents concurrent LanceDB access when multiple MCP clients or IDE sessions try to launch the server simultaneously.\n- **Sub-Second Hybrid Search:** Combined Vector + Tantivy FTS with **RAM-cached table connections** for instant retrieval.\n- **AST-Aware Chunking:** Uses `tree-sitter` to respect class and method boundaries in PHP, JS, and TS documentation.\n- **TurboQuant Acceleration:** Fast 4-bit quantized search for the `full-tq` profile, delivering 10x lower latency for large models.\n- **Neighborhood Retrieval:** Automatically fetches adjacent chunks (prev/next) to provide complete context for grounding.\n- **Macro-Context Summaries:** Injects file-level purpose summaries into every chunk using the `SummaryCache`.\n- **AI-Synthesized Answers:** Generates concise answers from search results with structured inline citations `[1]`, `[2]`.\n\n---\n\n## MCP Components\n\nThis server provides tools, prompts, and resources. See **[docs/mcp-components.md](docs/mcp-components.md)** for a full reference.\n\n### Primary Tools\n\n| Tool | Description |\n|---|---|\n| `search_plesk_unified` | Hybrid search with RRF and Cross-Encoder reranking. |\n| `get_file_content` | Retrieve the full content of a specific documentation file. |\n| `resolve_references` | Find all files referencing a specific symbol or topic. |\n| `refresh_knowledge` | Re-fetch sources and update the index (incremental). |\n| `trigger_index_sync` | Start a background indexing job. |\n| `daemon_health` | Check readiness, hardware acceleration (MPS/CUDA), and latency stats. |\n\n### Resources\n\n- `plesk://toc/api` - Table of Contents for API documentation.\n- `plesk://toc/cli` - Table of Contents for CLI reference.\n- `plesk://toc/guide` - Table of Contents for Extensions Guide.\n- `plesk://toc/php-stubs` - Hierarchical list of PHP classes.\n\n---\n\n## 🚀 Installation \u0026 Setup\n\nBecause this server is published to [PyPI](https://pypi.org/project/mcp-plesk-dev-docs/) and listed on the [MCP Registry](https://registry.modelcontextprotocol.io), you don't even need to clone the repository to run it!\n\n### Option 1: Run instantly via `uvx` (Recommended)\n\nYou can run or integrate the server in seconds.\n\n#### 1. Add to Claude Desktop\nAdd the server config to your `claude_desktop_config.json` (typically at `~/Library/Application Support/Claude/claude_desktop_config.json` on macOS, or `%APPDATA%\\Claude\\claude_desktop_config.json` on Windows):\n\n```json\n{\n  \"mcpServers\": {\n    \"plesk-dev-docs\": {\n      \"command\": \"uvx\",\n      \"args\": [\"mcp-plesk-dev-docs\"]\n    }\n  }\n}\n```\n\n#### 2. Configure in Cursor\nGo to **Settings \u003e Features \u003e MCP**, click **+ Add New MCP Server**:\n- **Name:** `plesk-dev-docs`\n- **Type:** `command`\n- **Command:** `uvx mcp-plesk-dev-docs`\n\n---\n\n### Option 2: Local Developer Setup (Manual Build)\n\nIf you want to modify the source code, run benchmarks, or manage database migrations:\n\n**Quick bootstrap (recommended):**\n```bash\ngit clone https://github.com/barateza/mcp-plesk-dev-docs.git\ncd mcp-plesk-dev-docs\n./install.sh          # Linux / macOS\n# powershell -ExecutionPolicy Bypass -File install.ps1   # Windows\n```\n\n**Manual setup:**\n```bash\ngit clone https://github.com/barateza/mcp-plesk-dev-docs.git\ncd mcp-plesk-dev-docs\nuv pip install -e \".[dev]\"\n```\n\n2. **Run Initial Indexing:**\n   Generate the offline vector database and full-text search indexes:\n   ```bash\n   uv run python -m mcp_plesk_dev_docs.server.main refresh_knowledge\n   ```\n\n3. **Start the Server:**\n   ```bash\n   uv run python -m mcp_plesk_dev_docs.server.main\n   ```\n\n---\n\n## Configuration\n\nSet environment variables in `.env`:\n\n```env\nPLESK_MODEL_PROFILE=light       # light | medium | full-tq\nPLESK_ENABLE_SAMPLING=true     # AI-Synthesized answers\nPLESK_DAEMON_AUTO_WARMUP=true  # Preload models on startup\nPLESK_INDEX_SUMMARIES=true     # Enable file-level summaries\nOPENROUTER_API_KEY=sk-or-v1-...\n```\n\n---\n\n## Documentation\n\n- **[docs/benchmarks.md](docs/benchmarks.md)** - Detailed latency and quality reports.\n- **[docs/mcp-components.md](docs/mcp-components.md)** - Full tool and resource reference.\n- **[docs/turboquant.md](docs/turboquant.md)** - 4-bit quantization internals.\n\n---\n\n## License\n\nMIT. See [LICENSE](LICENSE).\n\n## Ownership \u0026 Disclaimer\n\nThis is a personal project by Gilson Siqueira. It is not officially affiliated with, endorsed by, or supported by Plesk or WebPros International GmbH. Plesk is a trademark of WebPros International GmbH.\n\n### Important notice about Plesk-owned deliverables\n\nPortions of this repository were developed under contract for Plesk International GmbH (\"Plesk\") only if specifically identified as such. The MIT license above applies only to material the repository owner is authorized to license. Files or directories owned by Plesk, if any, are listed in [NOTICE](NOTICE). If you need assurance about licensing for a particular file, contact Plesk or seek legal counsel before relying on the MIT License for Plesk-owned files.\n\n---\n\n*Built to make Plesk extension development faster.*\n\n\u003c!-- mcp-name: io.github.barateza/mcp-plesk-dev-docs --\u003e\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbarateza%2Fmcp-plesk-dev-docs","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fbarateza%2Fmcp-plesk-dev-docs","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbarateza%2Fmcp-plesk-dev-docs/lists"}