blackmount-nlp-mcp
blackmount-nlp-mcp
面向 MCP 的 NLP 工具 — 零繁重依赖。 由 Blackmount 构建。
作为 FastMCP 服务器提供 45 种文本分析工具。无需 NLTK,无需 spaCy,无需 transformers。仅有一个依赖项 (mcp[cli]),NLP 代码量不到 50 KB,秒级就绪。需要 Python 3.10+。
为什么选择它
blackmount-nlp-mcp | NLTK | spaCy | transformers | |
Wheel 大小 | 42 KB | 1.5 MB | 6 MB+ (+ 模型) | 10 MB+ (+ 模型) |
直接依赖 | 1 | 多 | 多 | 多 |
分词 | ✅ | ✅ | ✅ | ✅ |
情感分析 | ✅ | ✅ | ❌ | ✅ |
可读性评分 | ✅ | ❌ | ❌ | ❌ |
关键词提取 | ✅ | ✅ | ❌ | ❌ |
文本相似度 | ✅ | ✅ | ✅ | ✅ |
语言检测 | ✅ (18 种语言) | ❌ | ❌ | ❌ |
所有功能均使用纯 Python 从零实现 — 包括 Porter 词干提取、TF-IDF、RAKE、Levenshtein 距离、VADER 风格情感分析、Flesch / Gunning Fog / Coleman-Liau / ARI / SMOG 可读性评分、抽取式摘要、语言检测 — 此外还内置了 2000+ 词的情感词典和 500+ 停用词列表,全部打包在内。
Related MCP server: LODA MCP Server
快速开始
pip install blackmount-nlp-mcpClaude Desktop
添加到你的配置文件中:
macOS:
~/Library/Application Support/Claude/claude_desktop_config.jsonWindows:
%APPDATA%\Claude\claude_desktop_config.jsonLinux:
~/.config/Claude/claude_desktop_config.json
{
"mcpServers": {
"nlp": {
"command": "blackmount-nlp-mcp"
}
}
}Cursor
添加到项目根目录的 .cursor/mcp.json 中:
{
"mcpServers": {
"nlp": {
"command": "blackmount-nlp-mcp"
}
}
}任意 MCP 客户端
服务器通过 stdio 运行。将你的客户端指向 blackmount-nlp-mcp 命令:
blackmount-nlp-mcp重启编辑器。现在所有 45 种 NLP 工具均已可用 — 只需用自然语言提问即可。
工具目录
分词 (4 种工具)
工具 | 描述 | 尝试提问 |
| 将文本拆分为单词,处理缩写和标点符号 | "Tokenize this paragraph into words" |
| 拆分为句子,处理常见缩写 | "Break this text into individual sentences" |
| 从词元列表中生成词级 n-grams | "Generate bigrams from these tokens" |
| 生成字符级 n-grams | "Get character trigrams for this word" |
可读性 (8 种工具)
所有可读性函数在输入为空或仅包含空格时返回 0.0。
工具 | 描述 | 尝试提问 |
| 0–100 易读性得分(越高越容易) | "Calculate the Flesch Reading Ease score" |
| 美国年级水平估算 | "What grade level is this written at?" |
| 基于复杂词比例的 Fog 指数 | "Calculate the Fog index for this text" |
| Coleman-Liau 年级水平指数 | "Get the Coleman-Liau score" |
| ARI 年级水平指数 | "What's the ARI for this document?" |
| SMOG 年级(推荐用于医疗文本) | "Calculate the SMOG grade for this document" |
| 任意单词的音节计数估算 | "How many syllables in 'extraordinary'?" |
| 一次调用获取所有可读性评分及纯英文标签 | "Give me a full readability report for this text" |
情感分析 (4 种工具)
工具 | 描述 | 尝试提问 |
| -1.0 到 +1.0 的复合情感得分 | "What's the sentiment of this customer review?" |
| 返回 | "Is this feedback positive or negative?" |
| 逐句情感分析 | "Show me the sentiment of each sentence" |
| 针对特定主题的情感分析 | "What's the sentiment around 'pricing' in these reviews?" |
关键词提取 (4 种工具)
工具 | 描述 | 尝试提问 |
| 语料库中的 TF-IDF 关键词排名 | "What are the key terms across these docs?" |
| RAKE 算法 — 短语级关键词提取 | "Extract the key phrases from this article" |
| 按频率排序的最高频词(排除停用词) | "What are the most common words in this text?" |
| 按频率排序的最高频 n-gram 短语 | "What two-word phrases appear most often?" |
文本相似度 (5 种工具)
工具 | 描述 | 尝试提问 |
| 词集重叠度,0–1 | "How similar are these two paragraphs?" |
| 词袋余弦相似度,0–1 | "Calculate cosine similarity between these texts" |
| Levenshtein 编辑距离 | "How many edits to turn 'kitten' into 'sitting'?" |
| 归一化到 0–1 的编辑距离 | "How different are these two strings?" |
| 两个字符串之间的 LCS 长度 | "What's the LCS length of these two strings?" |
文本清洗 (10 种工具)
工具 | 描述 | 尝试提问 |
| 去除 500+ 个英语停用词 | "Remove stopwords from this text" |
| 去除所有标点符号 | "Strip the punctuation" |
| 去除数字词元 | "Remove all numbers from this" |
| 去除 URL | "Clean out the URLs" |
| 去除电子邮件地址 | "Remove email addresses from this text" |
| 去除 HTML 标签 | "Strip the HTML from this content" |
| 合并并修剪空白字符 | "Normalize the whitespace" |
| 将文本转换为小写 | "Convert this to lowercase" |
| Porter 词干提取器(纯 Python,无 NLTK) | "Stem the word 'running'" |
| 可配置的多步清洗流水线 | "Clean this text: remove HTML, URLs, and stopwords" |
检测 (8 种工具)
工具 | 描述 | 尝试提问 |
| 识别 18 种支持的语言 | "What language is this text written in?" |
| 检测脚本类型:ASCII, Latin, Cyrillic, CJK, Arabic | "What script does this text use?" |
| 英语置信度得分,0–1 | "Is this text in English?" |
| 单词计数 | "How many words are in this?" |
| 句子计数 | "Count the sentences" |
| 段落计数 | "How many paragraphs?" |
| 平均单词长度(字符数) | "What's the average word length?" |
| 平均句子长度(单词数) | "How long are the sentences on average?" |
摘要 (2 种工具)
工具 | 描述 | 尝试提问 |
| 从文档中选择 N 个得分最高的句子 | "Summarize this article in 3 sentences" |
| 完整的文档统计:单词数、可读性、语言、阅读时间 | "Give me a statistical profile of this text" |
作为库使用
子模块可直接导入 — 无需 MCP 服务器:
from blackmount_nlp_mcp.sentiment import sentiment_score, sentiment_label
from blackmount_nlp_mcp.readability import reading_level
from blackmount_nlp_mcp.keywords import rake_keywords
text = "This product is absolutely amazing! The quality is excellent."
print(sentiment_score(text))
# 0.9285
print(sentiment_label(text))
# 'positive'
print(reading_level(text))
# {'grade_level': 12.39, 'label': 'college',
# 'flesch_reading_ease': 14.27, 'flesch_kincaid_grade': 12.39,
# 'gunning_fog': 19.58, 'coleman_liau': 10.94,
# 'automated_readability': 7.51, 'smog_grade': 11.21}
print(rake_keywords(text))
# [{'phrase': 'absolutely amazing', 'score': 4.0},
# {'phrase': 'product', 'score': 1.0},
# {'phrase': 'quality', 'score': 1.0},
# {'phrase': 'excellent', 'score': 1.0}]开发
git clone https://github.com/BlackMount-ai/blackmount-nlp-mcp
cd blackmount-nlp-mcp
pip install -e .
pytest tests/ -vBlackmount 生态系统
blackmount-nlp-mcp 由 Blackmount 构建 — 为 AI 从业者打造的工具。
blackmount-mcp — 作为 MCP 服务器提供浏览器记忆、AI 聊天搜索和会话分析功能。将其与 blackmount-nlp-mcp 搭配使用,可以分析你保存的对话:从聊天记录中提取关键词、评估 AI 回复的可读性、检测跨会话的情感趋势。
app.blackmount.ai — 完整的 Blackmount 平台。搜索、整理并分析你的 AI 工具产生的一切内容。
许可证
MIT
Maintenance
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceProvides intelligent summarization capabilities through a clean, extensible architecture. Mainly built for solving AI agents issues on big repositories, where large files can eat up the context window.2537MIT
- AlicenseNot gradedqualityDmaintenanceProvides token-efficient document search and retrieval for LLMs by returning relevant document sections within specified token budgets. It utilizes section-aware parsing and Bloom filter elimination to offer high-speed, zero-dependency access to large documents.MIT
- AlicenseAqualityAmaintenanceWeb content extraction for AI agents. 10 tools: scrape, crawl, map, batch, extract, summarize, diff, brand, search, research. Uses TLS fingerprinting to bypass anti-bot without a headless browser. Outputs LLM-optimized markdown with 67% fewer tokens than raw HTML.102,292AGPL 3.0
- AlicenseAqualityCmaintenanceSmart memory for AI agents. Solves the Karpathy problem: memories decay, topics are frequency-weighted, one-time questions don't become obsessions. 7 tools. Zero deps.4112MIT
Related MCP Connectors
Deterministic AI agent microtools, no accounts/API keys. fetch_extract: 98% token cut. 38 tools.
16 AI-native tools with dual SSE + streamable-http transport. Free tier available.
Honest library picks for coding agents in 25-360 tokens. Tells your agent what NOT to install.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/BlackMount-ai/blackmount-nlp-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server