mcp-local-rag 0.1.3 → 0.1.5

package/README.md

@@ -40,8 +40,8 @@ claude mcp add local-rag --scope user --env BASE_DIR=/path/to/your/documents --
 
 Restart your tool, then start using:
 ```
-"@mcp Ingest api-spec.pdf"
-"@mcp What does this document say about authentication?"
+"Ingest api-spec.pdf"
+"What does this document say about authentication?"
 ```
 
 That's it. No installation, no Docker, no complex setup.
@@ -80,23 +80,24 @@ All of this uses:
 
 The result: query responses typically under 3 seconds on a standard laptop, even with thousands of document chunks indexed.
 
-## Prerequisites
+## First Run
 
-You need Node.js 20 or higher. Check your version:
+On first launch, the embedding model downloads automatically from HuggingFace:
+- **Download size**: ~90MB (model files)
+- **Disk usage after caching**: ~120MB (includes ONNX runtime cache)
+- **Time**: 1-2 minutes on a decent connection
 
-```bash
-node --version
-```
+You'll see progress in the console. The model caches in `CACHE_DIR` (default: `./models/`) for offline use.
 
-If you need to install or update Node.js, visit [nodejs.org](https://nodejs.org/).
+**Offline Mode**: After first run, works completely offline—no internet required.
 
-That's all. When you configure the MCP server with `npx -y mcp-local-rag`, it automatically downloads and runs the package. No manual installation needed.
+## Security
 
-### First Run
+**Path Restriction**: This server only accesses files within your `BASE_DIR`. Any attempt to access files outside this directory (e.g., via `../` path traversal) will be rejected.
 
-On first launch, the embedding model downloads automatically from HuggingFace (about 90MB). This happens once—after that, it runs from your local cache.
+**Local Only**: All processing happens on your machine. No network requests are made after the initial model download.
 
-The download takes 1-2 minutes on a decent connection. You'll see progress in the console.
+**Model Verification**: The embedding model downloads from HuggingFace's official repository (`Xenova/all-MiniLM-L6-v2`). Verify integrity by checking the [official model card](https://huggingface.co/Xenova/all-MiniLM-L6-v2).
 
 ## Configuration
 
@@ -168,30 +169,31 @@ claude mcp add local-rag --scope user \
 
 ### Environment Variables
 
-**BASE_DIR** - Where your documents live. The server only accesses files in this directory (and subdirectories), preventing accidental access to system files. Defaults to your current working directory.
-
-**DB_PATH** - Where to store the vector database. Defaults to `./lancedb/` in your working directory. This directory can grow large if you index many documents.
-
-**CACHE_DIR** - Where Transformers.js caches the embedding model. Defaults to `./models/`. After the first download, the model stays here for offline use.
-
-**MODEL_NAME** - Which embedding model to use. Defaults to `Xenova/all-MiniLM-L6-v2`. Advanced users can try other models from HuggingFace, but they must be compatible with Transformers.js.
-
-**MAX_FILE_SIZE** - Maximum file size in bytes. Defaults to 104857600 (100MB). Larger files are rejected to prevent memory issues.
-
-**CHUNK_SIZE** - How many characters per chunk. Defaults to 512. Larger chunks give more context but slower processing.
-
-**CHUNK_OVERLAP** - How many characters overlap between chunks. Defaults to 100. This helps preserve context across chunk boundaries.
+| Variable | Default | Description | Valid Range |
+|----------|---------|-------------|-------------|
+| `BASE_DIR` | Current directory | Document root directory. Server only accesses files within this path (prevents accidental system file access). | Any valid path |
+| `DB_PATH` | `./lancedb/` | Vector database storage location. Can grow large with many documents. | Any valid path |
+| `CACHE_DIR` | `./models/` | Model cache directory. After first download, model stays here for offline use. | Any valid path |
+| `MODEL_NAME` | `Xenova/all-MiniLM-L6-v2` | HuggingFace model identifier. Must be Transformers.js compatible. See [available models](https://huggingface.co/models?library=transformers.js&pipeline_tag=feature-extraction&sort=trending). **Note:** Changing models requires re-ingesting all documents as embeddings from different models are incompatible. | HF model ID |
+| `MAX_FILE_SIZE` | `104857600` (100MB) | Maximum file size in bytes. Larger files rejected to prevent memory issues. | 1MB - 500MB |
+| `CHUNK_SIZE` | `512` | Characters per chunk. Larger = more context but slower processing. | 128 - 2048 |
+| `CHUNK_OVERLAP` | `100` | Overlap between chunks. Preserves context across boundaries. | 0 - (CHUNK_SIZE/2) |
 
 ## Usage
 
-Once configured, restart your MCP client. The server appears as available tools that your AI assistant can use.
+**After configuration**, restart your MCP client:
+- **Cursor**: Fully quit and relaunch (Cmd+Q on Mac, not just closing windows)
+- **Codex**: Restart the IDE/extension
+- **Claude Code**: No restart needed—changes apply immediately
+
+The server will appear as available tools that your AI assistant can use.
 
 ### Ingesting Documents
 
-**In Cursor**, use the MCP prefix to invoke tools:
+**In Cursor**, the Composer Agent automatically uses MCP tools when needed:
 
 ```
-"@mcp Ingest the document at /Users/me/docs/api-spec.pdf"
+"Ingest the document at /Users/me/docs/api-spec.pdf"
 ```
 
 **In Codex CLI**, the assistant automatically uses configured MCP tools when needed:
@@ -206,11 +208,7 @@ codex "Ingest the document at /Users/me/docs/api-spec.pdf into the RAG system"
 "Ingest the document at /Users/me/docs/api-spec.pdf"
 ```
 
-The tool uses relative paths from your BASE_DIR. So if BASE_DIR is `/Users/me/docs`, you can just say:
-
-```
-"@mcp Ingest api-spec.pdf"
-```
+**Path Requirements**: The server requires **absolute paths** to files. Your AI assistant will typically convert natural language requests into absolute paths automatically. The `BASE_DIR` setting restricts access to only files within that directory tree for security, but you must still provide the full path.
 
 The server:
 1. Validates the file exists and is under 100MB
@@ -226,9 +224,9 @@ This takes roughly 5-10 seconds per MB on a standard laptop. You'll see a confir
 Ask questions in natural language:
 
 ```
-"@mcp What does the API documentation say about authentication?"
-"@mcp Find information about rate limiting"
-"@mcp Search for error handling best practices"
+"What does the API documentation say about authentication?"
+"Find information about rate limiting"
+"Search for error handling best practices"
 ```
 
 The server:
@@ -241,7 +239,7 @@ Results include the text content, which file it came from, and a relevance score
 You can request more results:
 
 ```
-"@mcp Search for database optimization tips, return 10 results"
+"Search for database optimization tips, return 10 results"
 ```
 
 The limit parameter accepts 1-20 results.
@@ -251,7 +249,7 @@ The limit parameter accepts 1-20 results.
 See what's indexed:
 
 ```
-"@mcp List all ingested files"
+"List all ingested files"
 ```
 
 This shows each file's path, how many chunks it produced, and when it was ingested.
@@ -259,7 +257,7 @@ This shows each file's path, how many chunks it produced, and when it was ingest
 Check system status:
 
 ```
-"@mcp Show the RAG server status"
+"Show the RAG server status"
 ```
 
 This reports total documents, total chunks, current memory usage, and uptime.
@@ -269,7 +267,7 @@ This reports total documents, total chunks, current memory usage, and uptime.
 If you update a document, ingest it again:
 
 ```
-"@mcp Re-ingest api-spec.pdf with the latest changes"
+"Re-ingest api-spec.pdf with the latest changes"
 ```
 
 The server automatically deletes old chunks for that file before adding new ones. No duplicates, no stale data.
@@ -341,24 +339,41 @@ Each module has clear boundaries:
 
 ## Performance
 
-Measured on a MacBook Pro M1 (16GB RAM):
+**Test Environment**: MacBook Pro M1 (16GB RAM), tested with v0.1.3 on Node.js 22 (January 2025)
 
-**Query response time:** Average 1.2 seconds for 10,000 indexed chunks (5 results). Well under the 3-second target for p90.
+**Query Performance**:
+- Average: 1.2 seconds for 10,000 indexed chunks (5 results)
+- Target: p90 < 3 seconds ✓
 
-**Ingestion speed:** 10MB PDF processes in about 45 seconds. Breakdown:
-- PDF parsing: ~8 seconds
-- Text chunking: ~2 seconds
-- Embedding generation: ~30 seconds
-- Database insertion: ~5 seconds
+**Ingestion Speed** (10MB PDF):
+- Total: ~45 seconds
+  - PDF parsing: ~8 seconds (17%)
+  - Text chunking: ~2 seconds (4%)
+  - Embedding generation: ~30 seconds (67%)
+  - Database insertion: ~5 seconds (11%)
 
-**Memory usage:** Peak ~800MB when ingesting a 50MB file. Stays under 1GB as designed.
+**Memory Usage**:
+- Baseline: ~200MB idle
+- Peak: ~800MB when ingesting 50MB file
+- Target: < 1GB ✓
 
-**Concurrent queries:** Handles 5 parallel queries without degradation. LanceDB's async API allows non-blocking operations.
+**Concurrent Queries**: Handles 5 parallel queries without degradation. LanceDB's async API allows non-blocking operations.
 
-Your results will vary based on hardware, especially CPU speed (since embeddings run on CPU, not GPU).
+**Note**: Your results will vary based on hardware, especially CPU speed (embeddings run on CPU, not GPU).
 
 ## Troubleshooting
 
+### "No results found" when searching
+
+**Cause**: Documents must be ingested before searching.
+
+**Solution**:
+1. First ingest documents: `"Ingest /path/to/document.pdf"`
+2. Verify ingestion: `"List all ingested files"`
+3. Then search: `"Search for [your query]"`
+
+**Common mistake**: Trying to search immediately after configuration without ingesting any documents.
+
 ### "Model download failed"
 
 The embedding model downloads from HuggingFace on first run. If you're behind a proxy or firewall, you might need to configure network settings.
@@ -445,7 +460,20 @@ Cloud services (OpenAI, Pinecone, etc.) typically offer better accuracy and scal
 
 **What file formats are supported?**
 
-PDF, DOCX, TXT, and Markdown. For other formats, convert them first or open an issue requesting support.
+Currently supported:
+- **PDF**: `.pdf` (uses pdf-parse)
+- **Microsoft Word**: `.docx` (uses mammoth, not `.doc`)
+- **Plain Text**: `.txt`
+- **Markdown**: `.md`, `.markdown`
+
+**Not yet supported**:
+- Excel/CSV (`.xlsx`, `.csv`)
+- PowerPoint (`.pptx`)
+- Images with OCR (`.jpg`, `.png`)
+- HTML (`.html`)
+- Old Word documents (`.doc`)
+
+Want support for another format? [Open an issue](https://github.com/shinpr/mcp-local-rag/issues/new) with your use case.
 
 **Can I customize the embedding model?**
 
@@ -476,8 +504,6 @@ Contributions are welcome. Before submitting a PR:
 3. Add tests for new features
 4. Update documentation if you change behavior
 
-This project follows the [Conventional Commits](https://www.conventionalcommits.org/) standard for commit messages.
-
 ## License
 
 MIT License - see LICENSE file for details.

package/dist/embedder/index.js

@@ -2,7 +2,7 @@
 // Embedder implementation with Transformers.js
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.Embedder = exports.EmbeddingError = void 0;
-const transformers_1 = require("@xenova/transformers");
+const transformers_1 = require("@huggingface/transformers");
 // ============================================
 // Error Classes
 // ============================================

package/dist/embedder/index.js.map

@@ -1 +1 @@
-{"version":3,"file":"index.js","sourceRoot":"","sources":["../../src/embedder/index.ts"],"names":[],"mappings":";AAAA,+CAA+C;;;AAE/C,uDAAoD;AAkBpD,+CAA+C;AAC/C,gBAAgB;AAChB,+CAA+C;AAE/C;;GAEG;AACH,MAAa,cAAe,SAAQ,KAAK;IACvC,YACE,OAAe,EACC,KAAa;QAE7B,KAAK,CAAC,OAAO,CAAC,CAAA;QAFE,UAAK,GAAL,KAAK,CAAQ;QAG7B,IAAI,CAAC,IAAI,GAAG,gBAAgB,CAAA;IAC9B,CAAC;CACF;AARD,wCAQC;AAED,+CAA+C;AAC/C,iBAAiB;AACjB,+CAA+C;AAE/C;;;;;;;GAOG;AACH,MAAa,QAAQ;IAInB,YAAY,MAAsB;QAH1B,UAAK,GAAgD,IAAI,CAAA;QAI/D,IAAI,CAAC,MAAM,GAAG,MAAM,CAAA;IACtB,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,UAAU;QACd,IAAI,CAAC;YACH,+CAA+C;YAC/C,kBAAG,CAAC,QAAQ,GAAG,IAAI,CAAC,MAAM,CAAC,QAAQ,CAAA;YAEnC,OAAO,CAAC,GAAG,CAAC,yCAAyC,IAAI,CAAC,MAAM,CAAC,QAAQ,GAAG,CAAC,CAAA;YAC7E,OAAO,CAAC,GAAG,CAAC,4BAA4B,IAAI,CAAC,MAAM,CAAC,SAAS,MAAM,CAAC,CAAA;YACpE,IAAI,CAAC,KAAK,GAAG,MAAM,IAAA,uBAAQ,EAAC,oBAAoB,EAAE,IAAI,CAAC,MAAM,CAAC,SAAS,CAAC,CAAA;YACxE,OAAO,CAAC,GAAG,CAAC,qCAAqC,CAAC,CAAA;QACpD,CAAC;QAAC,OAAO,KAAK,EAAE,CAAC;YACf,MAAM,IAAI,cAAc,CACtB,kCAAmC,KAAe,CAAC,OAAO,EAAE,EAC5D,KAAc,CACf,CAAA;QACH,CAAC;IACH,CAAC;IAED;;;;;OAKG;IACH,KAAK,CAAC,KAAK,CAAC,IAAY;QACtB,IAAI,CAAC,IAAI,CAAC,KAAK,EAAE,CAAC;YAChB,MAAM,IAAI,cAAc,CAAC,uDAAuD,CAAC,CAAA;QACnF,CAAC;QAED,IAAI,CAAC;YACH,sCAAsC;YACtC,IAAI,IAAI,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;gBACtB,OAAO,IAAI,KAAK,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAA;YAC/B,CAAC;YAED,uEAAuE;YACvE,8FAA8F;YAC9F,MAAM,OAAO,GAAG,EAAE,OAAO,EAAE,MAAM,EAAE,SAAS,EAAE,IAAI,EAAE,CAAA;YACpD,MAAM,SAAS,GAAG,IAAI,CAAC,KAGa,CAAA;YACpC,MAAM,MAAM,GAAG,MAAM,SAAS,CAAC,IAAI,EAAE,OAAO,CAAC,CAAA;YAE7C,qCAAqC;YACrC,MAAM,SAAS,GAAG,KAAK,CAAC,IAAI,CAAC,MAAM,CAAC,IAAI,CAAC,CAAA;YACzC,OAAO,SAAS,CAAA;QAClB,CAAC;QAAC,OAAO,KAAK,EAAE,CAAC;YACf,MAAM,IAAI,cAAc,CACtB,iCAAkC,KAAe,CAAC,OAAO,EAAE,EAC3D,KAAc,CACf,CAAA;QACH,CAAC;IACH,CAAC;IAED;;;;;OAKG;IACH,KAAK,CAAC,UAAU,CAAC,KAAe;QAC9B,IAAI,CAAC,IAAI,CAAC,KAAK,EAAE,CAAC;YAChB,MAAM,IAAI,cAAc,CAAC,uDAAuD,CAAC,CAAA;QACnF,CAAC;QAED,IAAI,KAAK,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;YACvB,OAAO,EAAE,CAAA;QACX,CAAC;QAED,IAAI,CAAC;YACH,MAAM,UAAU,GAAe,EAAE,CAAA;YAEjC,6CAA6C;YAC7C,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,KAAK,CAAC,MAAM,EAAE,CAAC,IAAI,IAAI,CAAC,MAAM,CAAC,SAAS,EAAE,CAAC;gBAC7D,MAAM,KAAK,GAAG,KAAK,CAAC,KAAK,CAAC,CAAC,EAAE,CAAC,GAAG,IAAI,CAAC,MAAM,CAAC,SAAS,CAAC,CAAA;gBACvD,MAAM,eAAe,GAAG,MAAM,OAAO,CAAC,GAAG,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC,CAAC,CAAA;gBAChF,UAAU,CAAC,IAAI,CAAC,GAAG,eAAe,CAAC,CAAA;YACrC,CAAC;YAED,OAAO,UAAU,CAAA;QACnB,CAAC;QAAC,OAAO,KAAK,EAAE,CAAC;YACf,MAAM,IAAI,cAAc,CACtB,wCAAyC,KAAe,CAAC,OAAO,EAAE,EAClE,KAAc,CACf,CAAA;QACH,CAAC;IACH,CAAC;CACF;AAlGD,4BAkGC"}
+{"version":3,"file":"index.js","sourceRoot":"","sources":["../../src/embedder/index.ts"],"names":[],"mappings":";AAAA,+CAA+C;;;AAE/C,4DAAyD;AAkBzD,+CAA+C;AAC/C,gBAAgB;AAChB,+CAA+C;AAE/C;;GAEG;AACH,MAAa,cAAe,SAAQ,KAAK;IACvC,YACE,OAAe,EACC,KAAa;QAE7B,KAAK,CAAC,OAAO,CAAC,CAAA;QAFE,UAAK,GAAL,KAAK,CAAQ;QAG7B,IAAI,CAAC,IAAI,GAAG,gBAAgB,CAAA;IAC9B,CAAC;CACF;AARD,wCAQC;AAED,+CAA+C;AAC/C,iBAAiB;AACjB,+CAA+C;AAE/C;;;;;;;GAOG;AACH,MAAa,QAAQ;IAInB,YAAY,MAAsB;QAH1B,UAAK,GAAgD,IAAI,CAAA;QAI/D,IAAI,CAAC,MAAM,GAAG,MAAM,CAAA;IACtB,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,UAAU;QACd,IAAI,CAAC;YACH,+CAA+C;YAC/C,kBAAG,CAAC,QAAQ,GAAG,IAAI,CAAC,MAAM,CAAC,QAAQ,CAAA;YAEnC,OAAO,CAAC,GAAG,CAAC,yCAAyC,IAAI,CAAC,MAAM,CAAC,QAAQ,GAAG,CAAC,CAAA;YAC7E,OAAO,CAAC,GAAG,CAAC,4BAA4B,IAAI,CAAC,MAAM,CAAC,SAAS,MAAM,CAAC,CAAA;YACpE,IAAI,CAAC,KAAK,GAAG,MAAM,IAAA,uBAAQ,EAAC,oBAAoB,EAAE,IAAI,CAAC,MAAM,CAAC,SAAS,CAAC,CAAA;YACxE,OAAO,CAAC,GAAG,CAAC,qCAAqC,CAAC,CAAA;QACpD,CAAC;QAAC,OAAO,KAAK,EAAE,CAAC;YACf,MAAM,IAAI,cAAc,CACtB,kCAAmC,KAAe,CAAC,OAAO,EAAE,EAC5D,KAAc,CACf,CAAA;QACH,CAAC;IACH,CAAC;IAED;;;;;OAKG;IACH,KAAK,CAAC,KAAK,CAAC,IAAY;QACtB,IAAI,CAAC,IAAI,CAAC,KAAK,EAAE,CAAC;YAChB,MAAM,IAAI,cAAc,CAAC,uDAAuD,CAAC,CAAA;QACnF,CAAC;QAED,IAAI,CAAC;YACH,sCAAsC;YACtC,IAAI,IAAI,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;gBACtB,OAAO,IAAI,KAAK,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAA;YAC/B,CAAC;YAED,uEAAuE;YACvE,8FAA8F;YAC9F,MAAM,OAAO,GAAG,EAAE,OAAO,EAAE,MAAM,EAAE,SAAS,EAAE,IAAI,EAAE,CAAA;YACpD,MAAM,SAAS,GAAG,IAAI,CAAC,KAGa,CAAA;YACpC,MAAM,MAAM,GAAG,MAAM,SAAS,CAAC,IAAI,EAAE,OAAO,CAAC,CAAA;YAE7C,qCAAqC;YACrC,MAAM,SAAS,GAAG,KAAK,CAAC,IAAI,CAAC,MAAM,CAAC,IAAI,CAAC,CAAA;YACzC,OAAO,SAAS,CAAA;QAClB,CAAC;QAAC,OAAO,KAAK,EAAE,CAAC;YACf,MAAM,IAAI,cAAc,CACtB,iCAAkC,KAAe,CAAC,OAAO,EAAE,EAC3D,KAAc,CACf,CAAA;QACH,CAAC;IACH,CAAC;IAED;;;;;OAKG;IACH,KAAK,CAAC,UAAU,CAAC,KAAe;QAC9B,IAAI,CAAC,IAAI,CAAC,KAAK,EAAE,CAAC;YAChB,MAAM,IAAI,cAAc,CAAC,uDAAuD,CAAC,CAAA;QACnF,CAAC;QAED,IAAI,KAAK,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;YACvB,OAAO,EAAE,CAAA;QACX,CAAC;QAED,IAAI,CAAC;YACH,MAAM,UAAU,GAAe,EAAE,CAAA;YAEjC,6CAA6C;YAC7C,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,KAAK,CAAC,MAAM,EAAE,CAAC,IAAI,IAAI,CAAC,MAAM,CAAC,SAAS,EAAE,CAAC;gBAC7D,MAAM,KAAK,GAAG,KAAK,CAAC,KAAK,CAAC,CAAC,EAAE,CAAC,GAAG,IAAI,CAAC,MAAM,CAAC,SAAS,CAAC,CAAA;gBACvD,MAAM,eAAe,GAAG,MAAM,OAAO,CAAC,GAAG,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC,CAAC,CAAA;gBAChF,UAAU,CAAC,IAAI,CAAC,GAAG,eAAe,CAAC,CAAA;YACrC,CAAC;YAED,OAAO,UAAU,CAAA;QACnB,CAAC;QAAC,OAAO,KAAK,EAAE,CAAC;YACf,MAAM,IAAI,cAAc,CACtB,wCAAyC,KAAe,CAAC,OAAO,EAAE,EAClE,KAAc,CACf,CAAA;QACH,CAAC;IACH,CAAC;CACF;AAlGD,4BAkGC"}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "mcp-local-rag",
-  "version": "0.1.3",
+  "version": "0.1.5",
   "description": "Local RAG MCP Server - Easy-to-setup document search with minimal configuration",
   "main": "dist/index.js",
   "bin": {
@@ -33,7 +33,6 @@
     "watch": "tsx watch src/index.ts",
     "type-check": "tsc --noEmit",
     "test": "vitest run",
-    "test:ui": "vitest --ui",
     "test:coverage": "vitest run --coverage",
     "test:coverage:summary": "node scripts/show-coverage.js",
     "test:coverage:clean": "rm -rf coverage .vitest-cache",
@@ -55,10 +54,10 @@
     "prepare": "husky"
   },
   "dependencies": {
+    "@huggingface/transformers": "^3.7.6",
     "@lancedb/lancedb": "^0.22.2",
     "@langchain/textsplitters": "^1.0.0",
     "@modelcontextprotocol/sdk": "^1.20.2",
-    "@xenova/transformers": "^2.17.2",
     "langchain": "^1.0.2",
     "mammoth": "^1.11.0",
     "pdf-parse": "^2.4.5"
@@ -67,7 +66,6 @@
     "@biomejs/biome": "^1.9.4",
     "@types/node": "^20.0.0",
     "@vitest/coverage-v8": "^3.2.4",
-    "@vitest/ui": "^3.2.4",
     "c8": "^10.1.3",
     "husky": "^9.1.7",
     "jsdom": "^25.0.0",
@@ -81,7 +79,7 @@
     "vitest": "^3.2.4"
   },
   "engines": {
-    "node": "20"
+    "node": ">=20"
   },
   "lint-staged": {
     "src/**/*.{ts,tsx}": [