mcpField
M
MCP RAG Server

MCP RAG Server

Local
github.com/sylphlabMemory & Knowledge21Updated 16w ago

About MCP RAG Server

MCP RAG Server is an archived RAG project explicitly marked deprecated, with no further features, fixes, or support. Its README points to CodeRAG as the successor; the successor’s capabilities are not features of this repository.

Key features

  1. 01Archived and deprecated
  2. 02No further features or fixes
  3. 03README points to CodeRAG

Use cases

  1. 01Check the project’s deprecated status
  2. 02Find the successor named in the README

MCP support

The project was an MCP server; the README does not describe the specific tools exposed by the old server.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
"mcpServers": {
"rag-server-mcp-sylphlab": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"exec"
]
}
}
}
Run commanddocker run -i --rm exec

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
ociexec

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

⚠️ DEPRECATED - Use CodeRAG Instead ⚠️


🚀 This project has evolved into something much better!


New Project npm


👉 github.com/SylphxAI/coderag 👈


SOTA family roadmap: docs/roadmap/sota-family-roadmap.md.


🎉 Why CodeRAG is 10x Better

We completely rewrote this project from scratch. Here's what you get:

⚡ Lightning Fast
FeatureOld (rag-server-mcp)New (CodeRAG)
Startup10-30s (ChromaDB + Ollama)<1s (no external deps)
IndexingMinutes (embedding API calls)Seconds (TF-IDF + optional vectors)
Search~500ms (vector only)<50ms (hybrid search)
Memory500MB+ (ChromaDB)<100MB (SQLite)
🧠 Smarter Search
FeatureOldNew
TF-IDF❌✅ StarCoder2 tokenizer
Vector Search✅ Basic✅ Hybrid (TF-IDF + Vector)
Code Understanding❌ Generic✅ Code-aware tokenization
Incremental Updates❌ Full rebuild✅ Smart diff detection
🔧 Zero Dependencies
# Old way (painful)
docker-compose up -d      # Start ChromaDB
ollama pull nomic-embed   # Download model
# Wait 30 seconds...

# New way (instant)
npx @sylphx/coderag-mcp
# Done. That's it. 🎉
🏆 Battle-Tested Accuracy
  • Smoothed IDF - No term gets ignored (even common ones like function)
  • Logarithmic boosting - Stable ranking without score explosion
  • StarCoder2 tokenizer - 4.7MB model trained on code
  • Pre-computed magnitudes - Memory-efficient cosine similarity

🚀 Migrate in 30 Seconds

Step 1: Uninstall old
// Remove from your MCP config
{
  "mcpServers": {
    "rag-server": { ... }  // DELETE THIS
  }
}
Step 2: Install new
{
  "mcpServers": {
    "coderag": {
      "command": "npx",
      "args": ["-y", "@sylphx/coderag-mcp"]
    }
  }
}
Step 3: Enjoy 🎉

No Docker. No Ollama. No ChromaDB. Just works.


📊 Feature Comparison

Featurerag-server-mcpCodeRAG
External ServicesChromaDB + OllamaNone
Docker RequiredYesNo
Startup Time10-30s<1s
Search AlgorithmVector onlyHybrid (TF-IDF + Vector)
Code TokenizationGenericStarCoder2 (code-aware)
Incremental IndexNoYes
Memory Usage500MB+<100MB
Low Memory ModeNoYes (SQL-based)
Offline SupportNo (needs Ollama)Yes (TF-IDF)
Vector SearchRequiredOptional
Actively Maintained❌ No✅ Yes

🔗 Links


⚠️ This Repository is Archived

No new features. No bug fixes. No support.

Use CodeRAG instead.


Migrate Now



Made with ❤️ by Sylphx

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.