# Keyphrases

> Use this tool when you need to extract meaningful keyphrases from text documents, solving problems in content analysis, document summarization, and SEO keyword research. It takes in text documents and outputs extracted keyphrases, with customizable settings for stop word filtering and keyphrase counts. Ideal for applications requiring automated identification of key concepts from textual content, supporting multiple text formats and deployment modes.

Canonical page: https://skillsregistry.net/skills/ivanrublev-keyphrases  
JSON: https://api.skillsregistry.net/v1/skills/ivanrublev-keyphrases

## Description

A keyphrase extraction server that uses BERT-based sentence transformers and spaCy tokenization to identify and extract meaningful keyphrases from text documents. Built by Ivan Rublev, it leverages the KeyBERT library with multilingual MiniLM models for semantic understanding and supports customizable stop word filtering, configurable keyphrase counts, and GPU acceleration when available. The implementation includes comprehensive file access controls through directory allowlists, handles multiple text formats, and provides both CLI and HTTP server modes with Docker deployment support. Useful for content analysis, document summarization, SEO keyword research, or any application requiring automated extraction of key concepts from textual content.

## Trust

- **Trust score (0–1):** 0.50
- **Verification tier:** unverified

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** cloud-infra
- **Updated:** 2026-04-30

## Source

- **Source listing:** [PulseMCP](https://www.pulsemcp.com/servers/ivanrublev-keyphrases)
- **Repository:** <https://github.com/ivanrublev/keyphrases-mcp>

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "ivanrublev-keyphrases"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/ivanrublev-keyphrases` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/ivanrublev-keyphrases/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt