Tech Stack

Complete Code

#!/usr/bin/env python3 # blog_generator.py import os import json import argparse from datetime import datetime from openai import OpenAI client = OpenAI(api_key=os.getenv("DEEPSEEK_API_KEY"), base_url="https://api.deepseek.com") OUTLINE_PROMPT = """You are a senior technical blog author. Please generate a detailed article outline for the following topic. Topic: {topic} Writing style: {style} Target audience: {audience} Please generate a detailed outline with 5-8 sections, each with 2-3 sub-points.""" ARTICLE_PROMPT = """You are a senior technical blog author. Please write a complete Markdown blog post based on the outline. Topic: {topic} Outline: {outline} Writing style: {style} Target audience: {audience} Requirements: 1. Use Markdown format with appropriate heading levels 2. Include code examples (if applicable) 3. Content should be detailed, each section at least 150 words 4. Start with an engaging introduction 5. End with a summary and outlook 6. Do not use HTML tags, pure Markdown format""" SERIES_PROMPT = """You are a technical blog planner. Please generate a series article plan for the following topic. Topic: {topic} Number of series articles: {count} Please generate a JSON array, each element containing: - title: Article title - slug: URL-friendly title - description: Short description (within 20 characters) - order: Sequence number Return only the JSON array, no other content.""" def generate_outline(topic, style="Technical depth", audience="Developers"): response = client.chat.completions.create( model="deepseek-chat", messages=[{"role": "user", "content": OUTLINE_PROMPT.format( topic=topic, style=style, audience=audience )}], temperature=0.7, max_tokens=2048 ) return response.choices[0].message.content def generate_article(topic, outline, style="Technical depth", audience="Developers"): response = client.chat.completions.create( model="deepseek-chat", messages=[{"role": "user", "content": ARTICLE_PROMPT.format( topic=topic, outline=outline, style=style, audience=audience )}], temperature=0.7, max_tokens=8192 ) return response.choices[0].message.content def generate_series(topic, count=5): response = client.chat.completions.create( model="deepseek-chat", messages=[{"role": "user", "content": SERIES_PROMPT.format( topic=topic, count=count )}], temperature=0.7, max_tokens=2048 ) text = response.choices[0].message.content text = text.replace("```json", "").replace("```", "").strip() return json.loads(text) def create_front_matter(title, description, tags, platform="hugo"): date = datetime.now().strftime("%Y-%m-%d") if platform == "hugo": return f"""--- title: "{title}" date: {date} draft: false tags: [{', '.join([f'\"{t}\"' for t in tags])}] description: "{description}" categories: ["技术"] ---""" elif platform == "hexo": return f"""--- title: {title} date: {date} tags: [{', '.join(tags)}] description: {description} ---""" else: return f"""--- title: {title} date: {date} tags: [{', '.join(tags)}] description: {description} ---""" def save_article(content, filename, front_matter=""): full_content = f"{front_matter}\n\n{content}" if front_matter else content os.makedirs("output", exist_ok=True) filepath = os.path.join("output", filename) with open(filepath, "w", encoding="utf-8") as f: f.write(full_content) return filepath def main(): parser = argparse.ArgumentParser(description="DeepSeek Markdown 博客生成器") parser.add_argument("topic", help="文章主题") parser.add_argument("--style", "-s", default="技术深度", help="写作风格") parser.add_argument("--audience", "-a", default="开发者", help="目标读者") parser.add_argument("--tags", "-t", nargs="+", default=[], help="文章标签") parser.add_argument("--platform", "-p", default="hugo", choices=["hugo", "hexo", "jekyll", "vitepress"]) parser.add_argument("--series", type=int, help="生成系列文章规划（指定数量）") parser.add_argument("--desc", "-d", default="", help="文章描述") args = parser.parse_args() if args.series: print(f"正在生成「{args.topic}」系列文章规划 ({args.series} 篇)...") series = generate_series(args.topic, args.series) filepath = save_article(json.dumps(series, ensure_ascii=False, indent=2), "series_plan.json") print(f"系列规划已保存至: {filepath}") for item in series: print(f" [{item['order']}] {item['title']} - {item['description']}") return print(f"正在生成「{args.topic}」大纲...") outline = generate_outline(args.topic, args.style, args.audience) print("大纲:\n" + outline) print(f"\n正在生成完整文章...") article = generate_article(args.topic, outline, args.style, args.audience) slug = args.topic.lower().replace(" ", "-").replace("/", "-") filename = f"{slug}.md" description = args.desc or f"关于{args.topic}的深度技术文章" tags = args.tags or ["技术", "教程"] front_matter = create_front_matter(args.topic, description, tags, args.platform) filepath = save_article(article, filename, front_matter) word_count = len(article) print(f"\n文章已保存至: {filepath}") print(f"字数: {word_count}") if __name__ == "__main__": main()

运行方式

# 安装依赖 pip install openai # 生成单篇文章 python blog_generator.py "Python 异步编程最佳实践" --tags Python 异步 编程 # 指定写作风格和目标读者 python blog_generator.py "Kubernetes 入门指南" --style "通俗易懂" --audience "运维新手" # 生成系列文章规划 python blog_generator.py "微服务架构设计" --series 5 # 指定博客平台 python blog_generator.py "React Hooks 深度解析" --platform hexo

预期效果

• 输入主题自动生成结构化大纲和完整 Markdown 文章
• 自动生成 SEO 友好的 front matter（Hugo/Hexo/Jekyll/VitePress）
• 支持系列文章规划，一次性生成多篇文章的选题和描述
• 文章包含代码示例、引言、总结等完整结构

Tech Stack

Complete Code

#!/usr/bin/env python3 # email_reply.py import os import imaplib import smtplib import email from email.mime.text import MIMEText from email.header import decode_header from openai import OpenAI client = OpenAI(api_key=os.getenv("DEEPSEEK_API_KEY"), base_url="https://api.deepseek.com") # Email configuration IMAP_SERVER = os.getenv("IMAP_SERVER", "imap.gmail.com") IMAP_PORT = int(os.getenv("IMAP_PORT", "993")) SMTP_SERVER = os.getenv("SMTP_SERVER", "smtp.gmail.com") SMTP_PORT = int(os.getenv("SMTP_PORT", "587")) EMAIL_USER = os.getenv("EMAIL_USER") EMAIL_PASS = os.getenv("EMAIL_PASS") REPLY_TEMPLATES = { "Product Inquiry": """Thank you for your inquiry! Regarding {product}, here are the details: {details} If you have any other questions, feel free to contact us.""", "After-sales Support": """Thank you for contacting after-sales support. Regarding the issue you reported, we suggest: {solution} If you need further assistance, please provide more information.""", "Business Cooperation": """Thank you for your interest in cooperation! We are very happy to communicate further with you. {response} Looking forward to working with you!""", "Other": """Thank you for your email. {response} Best regards!""" } CLASSIFY_PROMPT = """You are an email classification assistant. Please classify the following email into one of these categories: - Product Inquiry: asking about product features, price, usage - After-sales Support: complaints, bug reports, returns/exchanges - Business Cooperation: cooperation intent, business negotiation - Spam: advertisements, scams, irrelevant content - Other: emails that cannot be categorized Return only the category name, no other content. Email subject: {subject} Email body: {body}""" REPLY_PROMPT = """You are a professional email reply assistant. Please generate a reply for the following email. Email category: {category} Email subject: {subject} Email body: {body} Reply template: {template} Please generate specific reply content, filling in the template variables. Requirements: 1. Reply professionally, politely, and concisely 2. Provide substantive answers to specific questions in the email 3. If the email information is insufficient, proactively ask for additional information needed Please return in JSON format, containing the variables needed to fill the template: {{"product": "...", "details": "...", "solution": "...", "response": "..."}}""" def decode_email_header(header): if not header: return "" decoded_parts = decode_header(header) subject = "" for part, charset in decoded_parts: if isinstance(part, bytes): subject += part.decode(charset or "utf-8", errors="ignore") else: subject += str(part) return subject def get_email_body(msg): body = "" if msg.is_multipart(): for part in msg.walk(): content_type = part.get_content_type() if content_type == "text/plain": try: body = part.get_payload(decode=True).decode("utf-8", errors="ignore") except: body = str(part.get_payload()) break else: try: body = msg.get_payload(decode=True).decode("utf-8", errors="ignore") except: body = str(msg.get_payload()) return body[:3000] def fetch_unread_emails(limit=10): mail = imaplib.IMAP4_SSL(IMAP_SERVER, IMAP_PORT) mail.login(EMAIL_USER, EMAIL_PASS) mail.select("inbox") _, data = mail.search(None, "UNSEEN") email_ids = data[0].split()[-limit:] emails = [] for eid in email_ids: _, msg_data = mail.fetch(eid, "(RFC822)") msg = email.message_from_bytes(msg_data[0][1]) subject = decode_email_header(msg["Subject"]) sender = msg["From"] body = get_email_body(msg) emails.append({"id": eid.decode(), "subject": subject, "sender": sender, "body": body}) mail.logout() return emails def classify_email(subject, body): response = client.chat.completions.create( model="deepseek-chat", messages=[{"role": "user", "content": CLASSIFY_PROMPT.format( subject=subject[:200], body=body[:2000] )}], temperature=0.1, max_tokens=50 ) return response.choices[0].message.content.strip() def generate_reply(category, subject, body): template = REPLY_TEMPLATES.get(category, REPLY_TEMPLATES["Other"]) response = client.chat.completions.create( model="deepseek-chat", messages=[{"role": "user", "content": REPLY_PROMPT.format( category=category, subject=subject[:200], body=body[:2000], template=template )}], temperature=0.3, max_tokens=2048 ) import json try: text = response.choices[0].message.content text = text.replace("```json", "").replace("```", "").strip() vars_dict = json.loads(text) except: vars_dict = {"response": "We have received your email and will reply to you as soon as possible."} return template.format(**{k: v for k, v in vars_dict.items() if k in template}) def send_reply(to_address, subject, body): msg = MIMEText(body, "plain", "utf-8") msg["Subject"] = f"Re: {subject}" msg["From"] = EMAIL_USER msg["To"] = to_address server = smtplib.SMTP(SMTP_SERVER, SMTP_PORT) server.starttls() server.login(EMAIL_USER, EMAIL_PASS) server.send_message(msg) server.quit() print(f"Reply sent to: {to_address}") def main(): print("Checking unread emails...") emails = fetch_unread_emails() if not emails: print("No unread emails") return print(f"Found {len(emails)} unread emails\n") for i, mail in enumerate(emails): print(f"\n{'='*50}") print(f"[{i+1}/{len(emails)}] From: {mail['sender']}") print(f"Subject: {mail['subject']}") category = classify_email(mail["subject"], mail["body"]) if category == "Spam": print(f"Category: {category} (skipped)") continue print(f"Category: {category}") reply = generate_reply(category, mail["subject"], mail["body"]) print(f"\nGenerated reply:\n{reply}") choice = input("\nSend this reply? (y=send / n=skip / e=edit): ").strip().lower() if choice == "y": send_reply(mail["sender"], mail["subject"], reply) elif choice == "e": custom_reply = input("Enter your reply: ") if custom_reply.strip(): send_reply(mail["sender"], mail["subject"], custom_reply) print("Custom reply sent") else: print("Skipped") if __name__ == "__main__": if not EMAIL_USER or not EMAIL_PASS: print("Please set environment variables EMAIL_USER and EMAIL_PASS") print("For Gmail, you need to use an app-specific password") exit(1) main()

How to Run

# Install dependencies pip install openai

Set Environment Variables (Gmail Example)

export EMAIL_USER="your-email@gmail.com"
export EMAIL_PASS="your-app-password"
export IMAP_SERVER="imap.gmail.com"
export SMTP_SERVER="smtp.gmail.com"

# Run
python email_reply.py

Expected Results

• Automatically detect unread emails and categorize them (product inquiry/after-sales/business/spam)
• Generate professional replies using corresponding templates based on category
• Human review step: confirm before sending, support editing custom replies
• Automatically filter spam to improve processing efficiency

Tech Stack

Full Code

# resume_optimizer.py import os import streamlit as st import fitz from openai import OpenAI client = OpenAI(api_key=os.getenv("DEEPSEEK_API_KEY"), base_url="https://api.deepseek.com") ANALYZE_PROMPT = """You are a senior HR and resume optimization expert. Please analyze the match between the following resume and job description. Resume content: {resume} Job description: {jd} Please analyze from the following dimensions and provide results in JSON format: 1. Skill match (0-100%) 2. Experience match (0-100%) 3. Overall match (0-100%) 4. List of matched skills 5. List of missing skills 6. Resume strengths (3-5 items) 7. Resume improvement suggestions (3-5 items) 8. ATS optimization suggestions (3-5 items) Return JSON format: {{ "skill_match": number, "experience_match": number, "overall_match": number, "matched_skills": ["skill1", "skill2"], "missing_skills": ["skill1", "skill2"], "strengths": ["strength1", "strength2"], "improvements": ["suggestion1", "suggestion2"], "ats_suggestions": ["suggestion1", "suggestion2"] }}""" OPTIMIZE_PROMPT = """You are a senior resume optimization expert. Please optimize the resume content based on the following analysis and suggestions. Original resume: {resume} Job description: {jd} Match analysis: {analysis} Optimization requirements: 1. Highlight skills and experience that match the JD 2. Use industry keywords (ATS-friendly) 3. Quantify achievements (use numbers and percentages) 4. Keep the original resume's factual information, do not fabricate experience 5. Optimize wording and layout structure Please output the optimized complete resume.""" def extract_pdf_text(pdf_file): doc = fitz.open(stream=pdf_file.read(), filetype="pdf") text = "" for page in doc: text += page.get_text() return text def analyze_resume(resume_text, jd_text): response = client.chat.completions.create( model="deepseek-chat", messages=[{"role": "user", "content": ANALYZE_PROMPT.format( resume=resume_text[:5000], jd=jd_text[:3000] )}], temperature=0.1, max_tokens=2048 ) import json content = response.choices[0].message.content content = content.replace("```json", "").replace("```", "").strip() return json.loads(content) def optimize_resume(resume_text, jd_text, analysis): response = client.chat.completions.create( model="deepseek-chat", messages=[{"role": "user", "content": OPTIMIZE_PROMPT.format( resume=resume_text[:5000], jd=jd_text[:3000], analysis=json.dumps(analysis, ensure_ascii=False, indent=2) )}], temperature=0.3, max_tokens=4096 ) return response.choices[0].message.content st.set_page_config(page_title="DeepSeek Resume Optimization", layout="wide") st.title("DeepSeek Resume Optimization Tool") col1, col2 = st.columns(2) with col1: st.subheader("Upload Resume")