Spaces:
Running
Running
Commit
·
928a62e
unverified
·
0
Parent(s):
initial
Browse files- .gitignore +10 -0
- .python-version +1 -0
- Dockerfile +36 -0
- README.md +0 -0
- main.py +6 -0
- pyproject.toml +10 -0
- server.py +159 -0
- uv.lock +0 -0
.gitignore
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Python-generated files
|
| 2 |
+
__pycache__/
|
| 3 |
+
*.py[oc]
|
| 4 |
+
build/
|
| 5 |
+
dist/
|
| 6 |
+
wheels/
|
| 7 |
+
*.egg-info
|
| 8 |
+
|
| 9 |
+
# Virtual environments
|
| 10 |
+
.venv
|
.python-version
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
3.13
|
Dockerfile
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FROM python:3.13-slim
|
| 2 |
+
|
| 3 |
+
# Install uv
|
| 4 |
+
COPY --from=ghcr.io/astral-sh/uv:latest /uv /uvx /bin/
|
| 5 |
+
|
| 6 |
+
# Change the working directory to the `app` directory
|
| 7 |
+
WORKDIR /app
|
| 8 |
+
|
| 9 |
+
# Install dependencies (separate layer for better caching)
|
| 10 |
+
RUN --mount=type=cache,target=/root/.cache/uv \
|
| 11 |
+
--mount=type=bind,source=uv.lock,target=uv.lock \
|
| 12 |
+
--mount=type=bind,source=pyproject.toml,target=pyproject.toml \
|
| 13 |
+
uv sync --locked --no-install-project
|
| 14 |
+
|
| 15 |
+
# Copy the project into the image
|
| 16 |
+
COPY . /app
|
| 17 |
+
|
| 18 |
+
# Sync the project
|
| 19 |
+
RUN --mount=type=cache,target=/root/.cache/uv \
|
| 20 |
+
uv sync --locked
|
| 21 |
+
|
| 22 |
+
# Expose port 7860 for Hugging Face Spaces
|
| 23 |
+
EXPOSE 7860
|
| 24 |
+
|
| 25 |
+
# Set port environment variable for Hugging Face Spaces
|
| 26 |
+
ENV PORT=7860
|
| 27 |
+
|
| 28 |
+
# Set UV cache directory to writable location for non-root user
|
| 29 |
+
ENV UV_CACHE_DIR=/app/.cache/uv
|
| 30 |
+
|
| 31 |
+
# Create cache directory with proper permissions
|
| 32 |
+
RUN mkdir -p /app/.cache/uv && chmod -R 777 /app/.cache
|
| 33 |
+
|
| 34 |
+
# Run the server
|
| 35 |
+
CMD ["uv", "run", "server.py"]
|
| 36 |
+
|
README.md
ADDED
|
File without changes
|
main.py
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
def main():
|
| 2 |
+
print("Hello from sumit-mcp-server!")
|
| 3 |
+
|
| 4 |
+
|
| 5 |
+
if __name__ == "__main__":
|
| 6 |
+
main()
|
pyproject.toml
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[project]
|
| 2 |
+
name = "sumit-mcp-server"
|
| 3 |
+
version = "0.1.0"
|
| 4 |
+
description = "Add your description here"
|
| 5 |
+
readme = "README.md"
|
| 6 |
+
requires-python = ">=3.13"
|
| 7 |
+
dependencies = [
|
| 8 |
+
"bs4>=0.0.2",
|
| 9 |
+
"fastmcp>=2.12.4",
|
| 10 |
+
]
|
server.py
ADDED
|
@@ -0,0 +1,159 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
from fastmcp import FastMCP
|
| 3 |
+
import os
|
| 4 |
+
import requests
|
| 5 |
+
from bs4 import BeautifulSoup
|
| 6 |
+
from urllib.parse import urljoin, urlparse
|
| 7 |
+
import json
|
| 8 |
+
from typing import List, Dict, Optional
|
| 9 |
+
|
| 10 |
+
mcp = FastMCP("Tatva Website Data Access 🌐")
|
| 11 |
+
|
| 12 |
+
BASE_URL = "https://tatva.sumityadav.com.np"
|
| 13 |
+
|
| 14 |
+
def _fetch_page_content(url: str) -> Dict[str, str]:
|
| 15 |
+
"""Helper function to fetch and parse content from a specific page URL"""
|
| 16 |
+
try:
|
| 17 |
+
response = requests.get(url, timeout=10)
|
| 18 |
+
response.raise_for_status()
|
| 19 |
+
|
| 20 |
+
soup = BeautifulSoup(response.content, 'html.parser')
|
| 21 |
+
|
| 22 |
+
# Extract title
|
| 23 |
+
title = soup.find('title')
|
| 24 |
+
title_text = title.get_text().strip() if title else "No title found"
|
| 25 |
+
|
| 26 |
+
# Extract main content (try common content selectors)
|
| 27 |
+
content_selectors = ['main', 'article', '.content', '.post-content', '.entry-content', 'body']
|
| 28 |
+
content_text = ""
|
| 29 |
+
|
| 30 |
+
for selector in content_selectors:
|
| 31 |
+
content = soup.select_one(selector)
|
| 32 |
+
if content:
|
| 33 |
+
# Remove script and style elements
|
| 34 |
+
for script in content(["script", "style"]):
|
| 35 |
+
script.decompose()
|
| 36 |
+
content_text = content.get_text().strip()
|
| 37 |
+
break
|
| 38 |
+
|
| 39 |
+
if not content_text:
|
| 40 |
+
content_text = soup.get_text().strip()
|
| 41 |
+
|
| 42 |
+
return {
|
| 43 |
+
"url": url,
|
| 44 |
+
"title": title_text,
|
| 45 |
+
"content": content_text[:5000], # Limit content length
|
| 46 |
+
"status": "success"
|
| 47 |
+
}
|
| 48 |
+
except Exception as e:
|
| 49 |
+
return {
|
| 50 |
+
"url": url,
|
| 51 |
+
"title": "",
|
| 52 |
+
"content": "",
|
| 53 |
+
"status": f"error: {str(e)}"
|
| 54 |
+
}
|
| 55 |
+
|
| 56 |
+
@mcp.tool
|
| 57 |
+
def get_page_content(url: str) -> Dict[str, str]:
|
| 58 |
+
"""Fetch and parse content from a specific page URL"""
|
| 59 |
+
return _fetch_page_content(url)
|
| 60 |
+
|
| 61 |
+
@mcp.tool
|
| 62 |
+
def get_homepage_content() -> Dict[str, str]:
|
| 63 |
+
"""Get content from the homepage"""
|
| 64 |
+
return _fetch_page_content(BASE_URL)
|
| 65 |
+
|
| 66 |
+
@mcp.tool
|
| 67 |
+
def get_about_page() -> Dict[str, str]:
|
| 68 |
+
"""Get content from the about page"""
|
| 69 |
+
return _fetch_page_content(f"{BASE_URL}/about/")
|
| 70 |
+
|
| 71 |
+
@mcp.tool
|
| 72 |
+
def get_post_content(post_path: str) -> Dict[str, str]:
|
| 73 |
+
"""Get content from a specific blog post by providing the post path (e.g., '2025/09/22/advance_scanner/')"""
|
| 74 |
+
url = f"{BASE_URL}/posts/{post_path}"
|
| 75 |
+
return _fetch_page_content(url)
|
| 76 |
+
|
| 77 |
+
@mcp.tool
|
| 78 |
+
def get_all_posts_summary() -> List[Dict[str, str]]:
|
| 79 |
+
"""Get a summary of all available blog posts"""
|
| 80 |
+
posts = [
|
| 81 |
+
{"path": "2025/09/22/advance_scanner/", "title": "Advance Scanner"},
|
| 82 |
+
{"path": "2025/09/16/layers-travel/", "title": "Layers Travel"},
|
| 83 |
+
{"path": "2025/09/13/The Legend of Jimutavahana: A Detailed Story of Sacrifice and Compassion/", "title": "The Legend of Jimutavahana"},
|
| 84 |
+
{"path": "2025/09/13/mcp-proxy-sse-to-sse/", "title": "MCP Proxy SSE to SSE"},
|
| 85 |
+
{"path": "2025/09/05/embeddinggemma-300m/", "title": "EmbeddingGemma 300M"},
|
| 86 |
+
{"path": "2025/09/02/Ten Mahavidyas/", "title": "Ten Mahavidyas"},
|
| 87 |
+
{"path": "2025/08/06/gpt-oss/", "title": "GPT OSS"},
|
| 88 |
+
{"path": "2025/08/10/Kakbhushundi and Thoth/", "title": "Kakbhushundi and Thoth"},
|
| 89 |
+
{"path": "2025/07/29/lagpachai-maithili-festival-types-of-nagas/", "title": "Lagpachai Maithili Festival Types of Nagas"},
|
| 90 |
+
{"path": "2025/07/23/The Churia Forests/", "title": "The Churia Forests"},
|
| 91 |
+
{"path": "2025/07/17/The Brain's Thinking Cycle: From Thought to Reply/", "title": "The Brain's Thinking Cycle"},
|
| 92 |
+
{"path": "2025/07/14/clock/", "title": "Clock"},
|
| 93 |
+
{"path": "2025/07/13/WanderingMind/", "title": "Wandering Mind"},
|
| 94 |
+
{"path": "2025/07/07/love/", "title": "Love"},
|
| 95 |
+
{"path": "2025/07/06/Go:Backend Programming For ML Service/", "title": "Go: Backend Programming For ML Service"},
|
| 96 |
+
{"path": "2025/01/07/tatva-development-journey/", "title": "Tatva Development Journey"},
|
| 97 |
+
{"path": "2025/06/02/agentsAI/", "title": "Agents AI"},
|
| 98 |
+
{"path": "2025/06/02/bio/", "title": "Bio"},
|
| 99 |
+
{"path": "2025/06/03/tatva/", "title": "Tatva"},
|
| 100 |
+
{"path": "2024/06/01/agents/", "title": "Agents"},
|
| 101 |
+
{"path": "2024/05/30/lines/", "title": "Lines"},
|
| 102 |
+
{"path": "2024/01/21/life/", "title": "Life"},
|
| 103 |
+
{"path": "2025/05/30/familyTree/", "title": "Family Tree"},
|
| 104 |
+
{"path": "2021/10/07/dashain/", "title": "Dashain"},
|
| 105 |
+
{"path": "2025/09/03/moments/", "title": "Moments"}
|
| 106 |
+
]
|
| 107 |
+
return posts
|
| 108 |
+
|
| 109 |
+
@mcp.tool
|
| 110 |
+
def search_posts_by_keyword(keyword: str) -> List[Dict[str, str]]:
|
| 111 |
+
"""Search for posts containing a specific keyword in title or path"""
|
| 112 |
+
all_posts = get_all_posts_summary()
|
| 113 |
+
keyword_lower = keyword.lower()
|
| 114 |
+
|
| 115 |
+
matching_posts = []
|
| 116 |
+
for post in all_posts:
|
| 117 |
+
if keyword_lower in post["title"].lower() or keyword_lower in post["path"].lower():
|
| 118 |
+
matching_posts.append(post)
|
| 119 |
+
|
| 120 |
+
return matching_posts
|
| 121 |
+
|
| 122 |
+
@mcp.tool
|
| 123 |
+
def get_posts_by_year(year: str) -> List[Dict[str, str]]:
|
| 124 |
+
"""Get all posts from a specific year"""
|
| 125 |
+
all_posts = get_all_posts_summary()
|
| 126 |
+
year_posts = [post for post in all_posts if post["path"].startswith(f"{year}/")]
|
| 127 |
+
return year_posts
|
| 128 |
+
|
| 129 |
+
@mcp.resource("tatva://homepage")
|
| 130 |
+
def get_homepage_resource():
|
| 131 |
+
"""Resource for homepage content"""
|
| 132 |
+
content = get_homepage_content()
|
| 133 |
+
return json.dumps(content, indent=2)
|
| 134 |
+
|
| 135 |
+
@mcp.resource("tatva://about")
|
| 136 |
+
def get_about_resource():
|
| 137 |
+
"""Resource for about page content"""
|
| 138 |
+
content = get_about_page()
|
| 139 |
+
return json.dumps(content, indent=2)
|
| 140 |
+
|
| 141 |
+
@mcp.resource("tatva://posts/{post_path}")
|
| 142 |
+
def get_post_resource(post_path: str):
|
| 143 |
+
"""Resource for specific post content"""
|
| 144 |
+
content = get_post_content(post_path)
|
| 145 |
+
return json.dumps(content, indent=2)
|
| 146 |
+
|
| 147 |
+
@mcp.prompt
|
| 148 |
+
def analyze_post_prompt(post_path: str) -> str:
|
| 149 |
+
"""Generate a prompt for analyzing a specific blog post"""
|
| 150 |
+
return f"Please analyze the following blog post from Tatva website and provide insights about its content, themes, and key takeaways. Post path: {post_path}"
|
| 151 |
+
|
| 152 |
+
@mcp.prompt
|
| 153 |
+
def compare_posts_prompt(post_path1: str, post_path2: str) -> str:
|
| 154 |
+
"""Generate a prompt for comparing two blog posts"""
|
| 155 |
+
return f"Please compare and contrast these two blog posts from Tatva website, highlighting similarities, differences, and unique perspectives. Post 1: {post_path1}, Post 2: {post_path2}"
|
| 156 |
+
|
| 157 |
+
if __name__ == "__main__":
|
| 158 |
+
port = int(os.environ.get("PORT", 7860))
|
| 159 |
+
mcp.run(transport="sse", host="0.0.0.0", port=port,path="/api/mcp/")
|
uv.lock
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|