io.github.Opafex由 GitHub 帳號 Opafex 發布v0.1.1更新於 Oct 26, 2026
Text extraction, keyword extraction, language detection, and chunking for RAG
本機 MCP 伺服器，可從 PDF、DOCX 及文字/程式碼檔案擷取文字，並進行關鍵字擷取、內容類型辨識與供 RAG 使用的文字切塊。
Desktop only，透過 STDIO。 STDIO 服務會啟動本機處理程序，因此需要 SourceWeft 桌面主機。
參照 儲存庫 中的啟動說明。
MCP server for universal text extraction, keyword extraction, language detection, and text chunking — designed for RAG pipelines.
Supports PDF, DOCX, CSV, JSON, Markdown, HTML, and code files. Zero platform dependencies.
Requires Node.js 20+.
No API keys needed — everything runs locally.
extract_textExtract text content from a file.
extract_keywordsExtract ranked keywords from text. Filters stop words, sorted by frequency.
detect_languageDetect whether content is code, natural language, or mixed.
chunk_textSplit text into overlapping chunks for RAG/embedding pipelines.
process_fileAll-in-one: extract text, detect content type, extract keywords, and chunk.
server_info (built-in)Returns server metadata, registered tools, and resources.
Found a bug or have a feature request? Open an issue.
MIT — see LICENSE.
來源：packages/file-processor/README.md，提交 f85607b
Extract text content from a file. Supports PDF, DOCX, CSV, JSON, Markdown, HTML, and code files. Pass file content as base64 (for binary formats like PDF/DOCX) or raw text (for text formats). Returns extracted text, word count, and processing time.
content必填fileType必填Extract ranked keywords from text. Returns keywords sorted by frequency, useful for tagging, search indexing, and topic identification.
text必填limitDetect whether content is code, natural language, or mixed. Optionally detects the programming language from file extension. Useful for routing content to appropriate processing pipelines.
text必填fileNameSplit text into overlapping chunks suitable for RAG/embedding pipelines. Configurable chunk size and overlap. Returns chunks with positional metadata.
text必填overlapchunkSizeAll-in-one file processing: extract text, detect content type, extract keywords, and chunk for RAG. Returns the complete processed result in a single call.
content必填fileType必填chunkSizechunkOverlapkeywordLimitReturns server metadata: name, version, registered tools, and registered resources. Useful for introspection and health checks.
無需輸入參數。
已滿足適用檢查項的 90%
該評分衡量此收錄條目的完整程度，依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價，也不表示伺服器安全或目前可用：可連通性是帶日期的檢查結果，評分也不包含安裝量或使用者評價。
請在註冊中心條目中宣告憑證請求標頭，或保持端點能夠回應探測，以便判定其認證方式。
npm install @opafex/mcp-file-processor{ "mcpServers": { "file-processor": { "command": "npx", "args": ["@opafex/mcp-file-processor"] } }}{ "success": true, "text": "Extracted text content here...", "wordCount": 142, "pageCount": 3, "processingTimeMs": 45}{ "success": true, "keywords": [ { "word": "javascript", "frequency": 8 }, { "word": "framework", "frequency": 4 }, { "word": "react", "frequency": 3 } ], "totalWords": 156}{ "success": true, "contentType": "code", "programmingLanguage": "typescript"}{ "success": true, "chunks": [ { "index": 0, "text": "...", "startOffset": 0, "endOffset": 1000, "wordCount": 167 }, { "index": 1, "text": "...", "startOffset": 800, "endOffset": 1800, "wordCount": 172 } ], "totalChunks": 5, "totalCharacters": 4200}npm run build # Compile TypeScriptnpm run dev # Run with tsxnpm test # Run testsnpm start # Run compiled serverimport { FileProcessorService } from '@opafex/mcp-file-processor/service';
const service = new FileProcessorService();
const result = await service.processFile( 'Hello world, this is a document.', '.txt', 500, // chunk size 100 // overlap);