silva-mcp
# SILVA-MCP
> A Model Context Protocol (MCP) Server for the [SILVA rRNA Database](https://www.arb-silva.de/)
>
> Version: 0.1.0 | MIT License
> **Disclaimer:** This is an **unofficial** community-driven project and is not affiliated with, endorsed by, or officially connected to SILVA, DSMZ, or the SILVA rRNA database team. SILVA is a registered trademark of its respective owners. For the official SILVA website, please visit [https://www.arb-silva.de/](https://www.arb-silva.de/).
[English](./README.en.md) | [中文文档](./README.zh.md)
## Quick Start
```bash
npm install
npm run build
node dist/index.js
```
Add to your MCP client config:
```json
{
"mcpServers": {
"silva-mcp": {
"command": "node",
"args": ["/path/to/SILVA-MCP/dist/index.js"]
}
}
}
```
## Tools
| Tool | Description |
|---|---|
| `silva_search` | Search rRNA sequences by organism, accession, taxonomy, etc. |
| `silva_sequence_detail` | Get detailed metadata for a sequence |
| `silva_testprobe` | Match a probe/primer against SILVA (TestProbe 3.0) |
| `silva_testprime` | Evaluate a primer pair's coverage (TestPrime 1.0) |
| `silva_browser` | Get the SILVA browser URL for a sequence |
| `silva_list_databases` | List available databases and taxonomies |
TDQS
Scored across 6 tools
Each tool has a distinct purpose: search vs. detail vs. database listing vs. two analysis jobs (TestProbe for single probes, TestPrime for primer pairs) are clearly separated. The only mild overlap is between silva_browser (interactive taxonomy view) and silva_sequence_detail (detailed metadata), but descriptions make the distinction workable.
All tools use a consistent snake_case verb/noun pattern with a uniform silva_ prefix (silva_search, silva_sequence_detail, silva_list_databases). The pattern is predictable throughout.
Six tools is well-scoped for a database query and analysis service, with no redundant or filler entries. Each tool earns its place covering retrieval, detail, listing, browsing, and two distinct analysis workflows.
The surface covers search, detail retrieval, database listing, browser access, and the two main SILVA analysis jobs, which is a solid lifecycle for this domain. Minor gaps exist around bulk sequence export/download or retrieval by sequence rather than ID, but agents can work around these.