MCP Catalogs
Home

narsil-mcp

by postrv·151·Score 51

Rust MCP server providing 90+ code intelligence tools across 32 languages with neural search and security analysis.

developer-toolsai-llmsecurity
19
Forks
1
Open issues
this month
Last commit
2d ago
Indexed

Overview

narsil-mcp is a comprehensive Rust-based MCP server offering deep code intelligence capabilities. It features 90 specialized tools across 32 programming languages, with support for neural semantic search, taint analysis, call graph generation, and supply chain security features like SBOM generation. The server is privacy-first, works entirely offline, and supports both local and remote repository analysis. Its Rust implementation provides memory safety and performance benefits, with a minimal binary size of ~30MB.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

you:Code analysis and refactoring in large codebases
you:Security auditing and vulnerability scanning
you:Neural code search and similarity matching
you:What languages are supported by narsil-mcp?
you:Does narsil-mcp require internet connectivity?

When to choose this

Choose narsil-mcp when you need comprehensive code analysis across multiple languages with security scanning and neural search capabilities, especially if you're working with large codebases and want offline processing.

When NOT to choose this

Avoid narsil-mcp if you only need basic LSP functionality without the heavy code analysis tools, or if you need real-time collaboration features as it processes everything locally.

Tools this server exposes

12 tools extracted from the README
  • search_symbols

    Search for symbols in code repositories

  • call_graph

    Generate function call graphs for code analysis

  • security_scan

    Perform security vulnerability scanning on code

  • generate_sbom

    Generate Software Bill of Materials for dependencies

  • semantic_search

    Find similar code using neural embeddings

  • taint_analysis

    Track data flow paths in code to identify potential security issues

  • extract_dependencies

    Extract and analyze project dependencies

  • find_code_patterns

    Find specific code patterns across repositories

  • extract_documentation

    Extract documentation comments and docstrings

  • analyze_control_flow

    Analyze control flow graphs of functions

  • detect_dead_code

    Identify unused or dead code in repositories

  • file_metadata

    Extract metadata and structure information from files

Note: Inferred tool names and descriptions from feature descriptions, usage examples, and the broad functionality described in the README. The server mentions 90 tools total but doesn't list them all explicitly.

Comparable tools

semantic-mcpast-grep-mcpripgrep-mcptree-sitter-lsp

Installation

Installation

**macOS / Linux (Homebrew):**

brew tap postrv/narsil
brew install narsil-mcp

**Windows (Scoop):**

scoop bucket add narsil https://github.com/postrv/scoop-narsil
scoop install narsil-mcp

**Rust/Cargo (all platforms):**

cargo install narsil-mcp

**Node.js/npm (all platforms):**

npm install -g narsil-mcp

Claude Desktop Configuration

{
  "mcpServers": {
    "narsil-mcp": {
      "command": "narsil-mcp",
      "args": ["--repos", "~/projects"]
    }
  }
}

FAQ

What languages are supported by narsil-mcp?
narsil-mcp supports 32 programming languages including Rust, Python, JavaScript, TypeScript, Go, C/C++, Java, C#, and many others. See the README for the complete list with supported extensions.
Does narsil-mcp require internet connectivity?
No, narsil-mcp is designed to work fully offline. All analysis happens locally on your machine. Neural search features require internet access only if using cloud-based embedding services.

On Hacker News

Recent discussion from the developer community.

Compare narsil-mcp with

GitHub →

Last updated · Auto-generated from public README + GitHub signals.