Merge branch 'main' of https://git.mamahaha.work/sangge/BackDoorBuster into feature/pyc-detection

MANIFEST.in

@@ -0,0 +1,2 @@
+include README.md
+include LICENSE

README.md

@@ -1,6 +1,7 @@
 # BackDoorBuster
 
 ![BackDoorBuster Banner](./banner.webp)
+
 ## 项目背景
 
 随着网络安全威胁的增加，恶意软件和后门的检测成为了保护个人和组织数据安全的重要任务。后门通常被隐藏在合法软件中，给黑客提供远程控制目标系统的能力。本项目旨在开发一个工具，能够有效识别和评估潜在的后门风险。
@@ -17,21 +18,66 @@
 - **报告生成**: 自动生成详细的检测报告，列出所有发现的敏感操作和对应的风险等级。
 - **持续更新与维护**: 随着新的后门技术和检测方法的出现，持续更新正则表达式库和评级标准。
 
+## 打包
+
+### pip
+
+#### 打包命令
+
+```bash
+pip install wheel
+python setup.py sdist bdist_wheel
+```
+
+执行上述命令后，会在 dist 目录下生成 .tar.gz 和 .whl 文件。
+
+#### 本地安装
+
+- 安装 .whl 文件：
+
+``` bash
+pip install dist/backdoor_buster-0.1.0-py3-none-any.whl
+```
+
+- 安装 .tar.gz 文件：
+
+``` bash
+pip install dist/backdoor_buster-0.1.0.tar.gz
+```
+
+#### 上传到 PyPI
+
+- 安装 twine：
+
+``` bash
+pip install twine
+```
+
+- 使用 twine 上传包到 PyPI：
+
+``` bash
+twine upload dist/*
+```
+
+需要提供 PyPI 的用户名和密码。如果没有 PyPI 账号，可以在 PyPI 注册。
+
+#### 使用 PyPI 安装
+
+包上传到 PyPI 后，可以通过以下命令安装：
+
+``` bash
+pip install backdoor_buster
+```
+
 ## 使用说明
 
-1. 安装依赖:
+1. 执行扫描:
-
-   ```bash
-   pip install -r requirements.txt
-   ```
-
-2. 执行扫描:
 
     ```bash
-    python scan.py <project_directory>
+    python -m detection <project_directory> -o <path> -m <mode>
     ```
 
-3. 查看报告:
+2. 查看报告:
 
     报告将以文本形式输出在控制台，并可选择输出到指定文件。
 

detection/__main__.py

@@ -0,0 +1,426 @@
+import os
+from typing import Dict, List, Tuple, Optional
+from reportlab.lib.pagesizes import letter
+from reportlab.lib.styles import getSampleStyleSheet
+from reportlab.platypus import Paragraph, Spacer, SimpleDocTemplate
+from .Regexdetection import find_dangerous_functions
+from .GPTdetection import detectGPT
+from .pyc_detection import disassemble_pyc
+from .utils import *
+import sys
+from colorama import init, Fore, Style
+
+SUPPORTED_EXTENSIONS = {".py", ".js", ".cpp", ".pyc"}
+OUTPUT_FORMATS = ["html", "md", "txt", "pdf"]
+ORDERS = [
+    "__import__",
+    "system",
+    "exec",
+    "popen",
+    "eval",
+    "subprocess",
+    "__getattribute__",
+    "getattr",
+    "child_process",
+]
+
+# Initialize colorama
+init(autoreset=True)
+
+ORANGE = "\033[38;5;214m"
+CYAN = Fore.CYAN
+
+
+def supports_color() -> bool:
+    """
+    Checks if the running terminal supports color output.
+
+    Returns:
+        bool: True if the terminal supports color, False otherwise.
+    """
+    # Windows support
+    if sys.platform == "win32":
+        return True
+    # Check if output is a TTY (terminal)
+    if hasattr(sys.stdout, "isatty") and sys.stdout.isatty():
+        return True
+    return False
+
+
+def supports_emoji() -> bool:
+    """
+    Checks if the running terminal supports emoji output.
+
+    Returns:
+        bool: True if the terminal supports emoji, False otherwise.
+    """
+    # This is a simple check. Modern terminals typically support emoji.
+    return sys.platform != "win32" or os.getenv("WT_SESSION") is not None
+
+
+def highlight_orders(line: str, risk_level: str, use_color: bool) -> str:
+    """
+    Highlights specific orders in the line based on risk level.
+
+    Args:
+        line (str): The line to highlight.
+        risk_level (str): The risk level of the line ("high", "medium", "low").
+        use_color (bool): Whether to use color for highlighting.
+
+    Returns:
+        str: The highlighted line.
+    """
+    risk_colors = {
+        "high": Fore.RED,
+        "medium": Fore.YELLOW,
+        "low": CYAN,
+    }
+    color = risk_colors.get(risk_level, Fore.WHITE) if use_color else ""
+    reset = Style.RESET_ALL if use_color else ""
+
+    for order in ORDERS:
+        line = line.replace(order, f"{color}{order}{reset}")
+    return line
+
+
+def generate_text_content(results: Dict[str, List[Tuple[int, str]]]) -> str:
+    """
+    Generates a formatted text report for security analysis results.
+
+    Args:
+        results (Dict[str, List[Tuple[int, str]]]): The security analysis results categorized by risk levels.
+
+    Returns:
+        str: The formatted text report as a string.
+    """
+    use_color = supports_color()
+    use_emoji = supports_emoji()
+
+    text_output = "Security Analysis Report\n"
+    text_output += "=" * 30 + "\n\n"
+
+    for risk_level, entries in results.items():
+        if entries and risk_level != "none":
+            risk_color = (
+                {
+                    "high": Fore.RED,
+                    "medium": Fore.YELLOW,
+                    "low": Fore.GREEN,
+                }.get(risk_level, Fore.WHITE)
+                if use_color
+                else ""
+            )
+
+            risk_title = (
+                {
+                    "High": "👹",
+                    "Medium": "👾",
+                    "Low": "👻",
+                }
+                if use_emoji
+                else {
+                    "High": "",
+                    "Medium": "",
+                    "Low": "",
+                }
+            )
+
+            text_output += f"{risk_color}{risk_level.capitalize()} Risk{risk_title[risk_level.capitalize()]}:{Style.RESET_ALL if use_color else ''}\n"
+            text_output += "-" * (len(risk_level) + 6) + "\n"
+            for line_num, line in entries:
+                line = highlight_orders(line, risk_level, use_color)
+                line_text = f"{Style.RESET_ALL if use_color else ''} {Fore.GREEN if use_color else ''}{line_num}{Style.RESET_ALL if use_color else ''}: {line}{Style.RESET_ALL if use_color else ''}\n"
+                text_output += line_text
+            text_output += "\n"
+
+    return text_output
+
+
+def output_results(
+    results: Dict[str, List[Tuple[int, str]]],
+    output_format: str,
+    output_file: Optional[str] = None,
+) -> None:
+    """
+    Outputs the security analysis results in the specified format.
+
+    Args:
+        results (Dict[str, List[Tuple[int, str]]]): The security analysis results categorized by risk levels.
+        output_format (str): The format to output the results in. Supported formats: "pdf", "html", "md", "txt".
+        output_file (Optional[str]): The name of the file to save the output. If None, prints to the terminal.
+    """
+    OUTPUT_FORMATS = {"pdf", "html", "md", "txt"}
+
+    if output_file:
+        file_name, file_ext = os.path.splitext(output_file)
+        if output_format not in OUTPUT_FORMATS:
+            output_format = "txt"
+            output_file = f"{file_name}.txt"
+        results_dir = os.path.dirname(output_file)
+        if not os.path.exists(results_dir) and results_dir != "":
+            os.makedirs(results_dir)
+        if output_format == "pdf":
+            output_pdf(results, output_file)
+        elif output_format == "html":
+            output_html(results, output_file)
+        elif output_format == "md":
+            output_markdown(results, output_file)
+        else:  # Default to txt
+            output_text(results, output_file)
+    else:
+        # If no output file is specified, default to text output to the terminal.
+        txt_output = generate_text_content(results)
+        print(txt_output)
+
+
+def output_pdf(results: Dict[str, List[Tuple[int, str]]], file_name):
+    doc = SimpleDocTemplate(file_name, pagesize=letter)
+    story = []
+    styles = getSampleStyleSheet()
+
+    # Add the title centered
+    title_style = styles["Title"]
+    title_style.alignment = 1  # Center alignment
+    title = Paragraph("Security Analysis Report", title_style)
+    story.append(title)
+    story.append(Spacer(1, 20))  # Space after title
+
+    # Add risk levels and entries
+    normal_style = styles["BodyText"]
+    for risk_level, entries in results.items():
+        if risk_level != "none":
+            story.append(
+                Paragraph(f"{risk_level.capitalize()} Risk:", styles["Heading2"])
+            )
+            for line_num, line in entries:
+                entry = Paragraph(f"Line {line_num}: {line}", normal_style)
+                story.append(entry)
+            story.append(Spacer(1, 12))  # Space between sections
+
+    doc.build(story)
+
+
+def output_html(results: Dict[str, List[Tuple[int, str]]], file_name=None):
+    """
+    Generates an HTML report for security analysis results.
+
+    Args:
+        results (Dict[str, List[Tuple[int, str]]]): The security analysis results categorized by risk levels.
+        file_name (Optional[str]): The name of the file to save the HTML output. If None, returns the HTML string.
+
+    Returns:
+        Optional[str]: The HTML string if file_name is None, otherwise None.
+    """
+    html_output = """
+    <html>
+    <head>
+        <meta charset="UTF-8">
+        <meta name="viewport" content="width=device-width, initial-scale=1.0">
+        <link rel="icon" href="https://s2.loli.net/2024/05/30/WDc6MekjbuCU9Qo.png">
+        <title>Security Analysis Report</title>
+        <style>
+            body {
+                background-image: url('https://s2.loli.net/2024/05/30/85Mv7leB2IRWNp6.jpg');
+                background-size: 100%, auto;
+                background-attachment: fixed;
+                font-family: Arial, sans-serif;
+            }
+            h1, h2 {
+                color: white;
+            }
+            ul {
+                list-style-type: none;
+                padding: 0;
+            }
+            li {
+                background: rgba(255, 255, 255, 0.8);
+                margin: 5px 0;
+                padding: 10px;
+                border-radius: 5px;
+            }
+        </style>
+    </head>
+    <body>
+        <h1>Security Analysis Report</h1>
+    """
+
+    for risk_level, entries in results.items():
+        if risk_level != "none":
+            risk_title = {
+                "High": f"<h2>{risk_level.capitalize()} Risk👹</h2><ul>",
+                "Medium": f"<h2>{risk_level.capitalize()} Risk👾</h2><ul>",
+                "Low": f"<h2>{risk_level.capitalize()} Risk👻</h2><ul>",
+            }
+            html_output += risk_title[risk_level.capitalize()]
+            for line_num, line in entries:
+                html_output += f"<li>{line_num}: {line}</li>"
+            html_output += "</ul>"
+
+    html_output += "</body></html>"
+
+    if file_name:
+        with open(file_name, "w", encoding="utf-8") as file:
+            file.write(html_output)
+        return None
+    else:
+        return html_output
+
+
+def output_markdown(results: Dict[str, List[Tuple[int, str]]], file_name=None):
+    """
+    Generates a Markdown report for security analysis results.
+
+    Args:
+        results (Dict[str, List[Tuple[int, str]]]): The security analysis results categorized by risk levels.
+        file_name (Optional[str]): The name of the file to save the Markdown output. If None, returns the Markdown string.
+
+    Returns:
+        Optional[str]: The Markdown string if file_name is None, otherwise None.
+    """
+    md_output = "# Security Analysis Report\n\n"
+
+    for risk_level, entries in results.items():
+        if risk_level != "none":
+            md_output += f"## {risk_level.capitalize()} Risk\n\n"
+            md_output += "| Line Number | Description |\n"
+            md_output += "|-------------|-------------|\n"
+            for line_num, line in entries:
+                md_output += f"| {line_num} | {line} |\n"
+            md_output += "\n"
+
+    if file_name:
+        with open(file_name, "w") as file:
+            file.write(md_output)
+        return None
+    else:
+        return md_output
+
+
+def output_text(results: Dict[str, List[Tuple[int, str]]], file_name=None):
+    """
+    Generates a plain text report for security analysis results.
+
+    Args:
+        results (Dict[str, List[Tuple[int, str]]]): The security analysis results categorized by risk levels.
+        file_name (Optional[str]): The name of the file to save the text output. If None, returns the text string.
+
+    Returns:
+        Optional[str]: The text string if file_name is None, otherwise None.
+    """
+    text_output = "Security Analysis Report\n"
+    text_output += "=" * len("Security Analysis Report") + "\n\n"
+
+    for risk_level, entries in results.items():
+        if risk_level != "none":
+            text_output += f"{risk_level.capitalize()} Risk:\n"
+            text_output += "-" * len(f"{risk_level.capitalize()} Risk:") + "\n"
+            for line_num, line in entries:
+                text_output += f"  Line {line_num}: {line}\n"
+            text_output += "\n"
+
+    if file_name:
+        with open(file_name, "w") as file:
+            file.write(text_output)
+        return None
+    else:
+        return text_output
+
+
+def checkModeAndDetect(mode: str, filePath: str, fileExtension: str, pycdc_addr: str):
+    # TODO:添加更多方式，这里提高代码的复用性和扩展性
+    if fileExtension == ".pyc":
+        # 反汇编pyc文件
+        file_content = disassemble_pyc(filePath, pycdc_addr)
+        if mode == "regex":
+            return find_dangerous_functions(file_content, fileExtension)
+        elif mode == "llm":
+            return detectGPT(file_content)
+        else:
+            return find_dangerous_functions(file_content, fileExtension)
+    else:
+        file_content = read_file_content(filePath)
+        if mode == "regex":
+            return find_dangerous_functions(file_content, fileExtension)
+        elif mode == "llm":
+            return detectGPT(file_content)
+        else:
+            return find_dangerous_functions(file_content, fileExtension)
+
+
+def process_path(
+    path: str, output_format: str, mode: str, pycdc_addr: str, output_file=None
+):
+    results = {"high": [], "medium": [], "low": [], "none": []}
+    if os.path.isdir(path):
+        for root, dirs, files in os.walk(path):
+            for file in files:
+                file_extension = os.path.splitext(file)[1]
+                if file_extension in SUPPORTED_EXTENSIONS:
+                    file_path = os.path.join(root, file)
+                    file_results = checkModeAndDetect(
+                        mode, file_path, file_extension, pycdc_addr
+                    )
+                    for key in file_results:
+                        if key != "none":  # Exclude 'none' risk level
+                            results[key].extend(
+                                [
+                                    (f"{file_path}: Line {line_num}", line)
+                                    for line_num, line in file_results[key]
+                                ]
+                            )
+    elif os.path.isfile(path):
+        file_extension = os.path.splitext(path)[1]
+        if file_extension in SUPPORTED_EXTENSIONS:
+            file_results = checkModeAndDetect(mode, path, file_extension, pycdc_addr)
+            for key in file_results:
+                if key != "none":  # Exclude 'none' risk level
+                    results[key].extend(
+                        [
+                            (f"{path}: Line {line_num}", line)
+                            for line_num, line in file_results[key]
+                        ]
+                    )
+        else:
+            print("Unsupported file type.")
+            return
+    else:
+        print("Invalid path.")
+        sys.exit(1)
+
+    output_results(results, output_format, output_file)
+
+
+def main():
+    import argparse
+
+    parser = argparse.ArgumentParser(
+        description="Backdoor detection tool.", prog="detection"
+    )
+    parser.add_argument("path", help="Path to the code to analyze")
+    parser.add_argument("-o", "--output", help="Output file path", default=None)
+    parser.add_argument(
+        "-m", "--mode", help="Mode of operation:[regex,llm]", default="regex"
+    )
+    parser.add_argument(
+        "-p", "--pycdc", help="Path to pycdc.exe to decompile", default=None
+    )
+    args = parser.parse_args()
+    output_format = "txt"  # Default output format
+    output_file = None
+    if args.output:
+        _, ext = os.path.splitext(args.output)
+        ext = ext.lower()
+        if ext in [".html", ".md", ".txt", ".pdf"]:
+            output_format = ext.replace(".", "")
+            output_file = args.output
+        else:
+            print(
+                "Your input file format was incorrect, the output has been saved as a TXT file."
+            )
+            output_file = args.output.rsplit(".", 1)[0] + ".txt"
+    # 如果未指定输出文件，则输出到 stdout；否则写入文件
+    process_path(args.path, output_format, args.mode, args.pycdc, output_file)
+
+
+if __name__ == "__main__":
+    main()

detection/backdoor_detection.py

@@ -1,210 +0,0 @@
-import os
-from typing import Dict, List, Tuple
-from reportlab.lib.pagesizes import letter
-from reportlab.lib.styles import getSampleStyleSheet
-from reportlab.platypus import Paragraph, Spacer, SimpleDocTemplate
-from .Regexdetection import find_dangerous_functions
-from .GPTdetection import detectGPT
-from .utils import *
-from .pyc_detection import disassemble_pyc
-import sys
-
-SUPPORTED_EXTENSIONS = {".py", ".js", ".cpp", ".pyc"}
-OUTPUT_FORMATS = ["html", "md", "txt", "pdf"]
-
-
-def generate_text_content(results):
-    text_output = "Security Analysis Report\n"
-    for risk_level, entries in results.items():
-        if entries and risk_level != "none":
-            text_output += f"{risk_level.capitalize()} Risk:\n"
-            for line_num, line in entries:
-                text_output += f"  Line {line_num}: {line}\n"
-    return text_output
-
-
-def output_results(results, output_format, output_file=None):
-    if output_file:
-        file_name = os.path.splitext(output_file)
-        if output_format not in OUTPUT_FORMATS:
-            output_format = "txt"
-            output_file = f"{file_name}.txt"
-        results_dir = os.path.dirname(output_file)
-        if not os.path.exists(results_dir):
-            os.makedirs(results_dir)
-        if output_format == "pdf":
-            output_pdf(results, output_file)
-        elif output_format == "html":
-            output_html(results, output_file)
-        elif output_format == "md":
-            output_markdown(results, output_file)
-        else:  # Default to txt
-            output_text(results, output_file)
-    else:
-        # If no output file is specified, default to text output to the terminal.
-        txt_output = generate_text_content(results)
-        print(txt_output)
-
-
-def output_pdf(results: Dict[str, List[Tuple[int, str]]], file_name):
-    doc = SimpleDocTemplate(file_name, pagesize=letter)
-    story = []
-    styles = getSampleStyleSheet()
-
-    # Add the title centered
-    title_style = styles["Title"]
-    title_style.alignment = 1  # Center alignment
-    title = Paragraph("Security Analysis Report", title_style)
-    story.append(title)
-    story.append(Spacer(1, 20))  # Space after title
-
-    # Add risk levels and entries
-    normal_style = styles["BodyText"]
-    for risk_level, entries in results.items():
-        if risk_level != "none":
-            story.append(
-                Paragraph(f"{risk_level.capitalize()} Risk:", styles["Heading2"])
-            )
-            for line_num, line in entries:
-                entry = Paragraph(f"Line {line_num}: {line}", normal_style)
-                story.append(entry)
-            story.append(Spacer(1, 12))  # Space between sections
-
-    doc.build(story)
-
-
-def output_html(results: Dict[str, List[Tuple[int, str]]], file_name=None):
-    html_output = "<html><head><title>Security Analysis Report</title></head><body>"
-    html_output += "<h1>Security Analysis Report</h1>"
-    for risk_level, entries in results.items():
-        if risk_level != "none":
-            html_output += f"<h2>{risk_level.capitalize()} Risk</h2><ul>"
-            for line_num, line in entries:
-                html_output += f"<li>{line_num}: {line}</li>"
-            html_output += "</ul>"
-    html_output += "</body></html>"
-    if file_name:
-        with open(file_name, "w") as file:
-            file.write(html_output)
-    else:
-        return html_output
-
-
-def output_markdown(results: Dict[str, List[Tuple[int, str]]], file_name=None):
-    md_output = "# Security Analysis Report\n"
-    for risk_level, entries in results.items():
-        if risk_level != "none":
-            md_output += f"## {risk_level.capitalize()} Risk\n"
-            for line_num, line in entries:
-                md_output += f"- {line_num}: {line}\n"
-    if file_name:
-        with open(file_name, "w") as file:
-            file.write(md_output)
-    else:
-        return md_output
-
-
-def output_text(results: Dict[str, List[Tuple[int, str]]], file_name=None):
-    text_output = "Security Analysis Report\n"
-    for risk_level, entries in results.items():
-        if risk_level != "none":
-            text_output += f"{risk_level.capitalize()} Risk:\n"
-            for line_num, line in entries:
-                text_output += f"  {line_num}: {line}\n"
-    if file_name:
-        with open(file_name, "w") as file:
-            file.write(text_output)
-    else:
-        return text_output
-
-
-def checkModeAndDetect(mode: str, filePath: str, fileExtension: str):
-    # TODO:添加更多方式，这里提高代码的复用性和扩展性
-    if fileExtension == ".pyc":
-        # 反汇编pyc文件
-        file_content = disassemble_pyc(filePath)
-        if mode == "regex":
-            return find_dangerous_functions(file_content, fileExtension)
-        elif mode == "llm":
-            return detectGPT(file_content)
-        else:
-            return find_dangerous_functions(file_content, fileExtension)
-    else:
-        file_content = read_file_content(filePath)
-        if mode == "regex":
-            return find_dangerous_functions(file_content, fileExtension)
-        elif mode == "llm":
-            return detectGPT(file_content)
-        else:
-            return find_dangerous_functions(file_content, fileExtension)
-
-
-def process_path(path: str, output_format: str, mode: str, output_file=None):
-    results = {"high": [], "medium": [], "low": [], "none": []}
-    if os.path.isdir(path):
-        for root, dirs, files in os.walk(path):
-            for file in files:
-                file_extension = os.path.splitext(file)[1]
-                if file_extension in SUPPORTED_EXTENSIONS:
-                    file_path = os.path.join(root, file)
-
-                    file_results = checkModeAndDetect(mode, file_path, file_extension)
-                    for key in file_results:
-                        if key != "none":  # Exclude 'none' risk level
-                            results[key].extend(
-                                [
-                                    (f"{file_path}: Line {line_num}", line)
-                                    for line_num, line in file_results[key]
-                                ]
-                            )
-    elif os.path.isfile(path):
-        file_extension = os.path.splitext(path)[1]
-        if file_extension in SUPPORTED_EXTENSIONS:
-            file_results = checkModeAndDetect(mode, path, file_extension)
-            for key in file_results:
-                if key != "none":  # Exclude 'none' risk level
-                    results[key].extend(
-                        [
-                            (f"{path}: Line {line_num}", line)
-                            for line_num, line in file_results[key]
-                        ]
-                    )
-        else:
-            print("Unsupported file type.")
-            return
-    else:
-        print("Invalid path.")
-        sys.exit(1)
-
-    output_results(results, output_format, output_file)
-
-
-def main():
-    import argparse
-
-    parser = argparse.ArgumentParser(description="Backdoor detection tool.")
-    parser.add_argument("path", help="Path to the code to analyze")
-    parser.add_argument("-o", "--output", help="Output file path", default=None)
-    parser.add_argument(
-        "-m", "--mode", help="Mode of operation:[regex,llm]", default="regex"
-    )
-    args = parser.parse_args()
-    output_format = "txt"  # Default output format
-    output_file = None
-    if args.output:
-        _, ext = os.path.splitext(args.output)
-        ext = ext.lower()
-        if ext in [".html", ".md", ".txt", ".pdf"]:
-            output_format = ext.replace(".", "")
-            output_file = args.output
-        else:
-            print(
-                "Your input file format was incorrect, the output has been saved as a TXT file."
-            )
-            output_file = args.output.rsplit(".", 1)[0] + ".txt"
-    # 如果未指定输出文件，则输出到 stdout；否则写入文件
-    process_path(args.path, output_format, args.mode, output_file)
-
-
-if __name__ == "__main__":
-    main()

detection/pyc_detection.py

@@ -1,9 +1,36 @@
 from typing import List, Tuple
 import uncompyle6
 import io
+import os
+import subprocess
 
 
-def disassemble_pyc(file_path: str) -> str:
+def run_pycdc(exe_path: str, pyc_file: str) -> str:
+    """
+    Executes pycdc.exe with the given .pyc file using a command line string and captures the output.
+
+    Args:
+        exe_path (str): Path to the pycdc.exe executable.
+        pyc_file (str): Path to the .pyc file to decompile.
+
+    Returns:
+        str: Output from pycdc.exe.
+    """
+    if not os.path.isfile(exe_path):
+        print(f"ERROR: The specified pycdc.exe path is not valid: {exe_path}")
+        print("Please check your pycdc path.")
+        exit(1)
+
+    command = f'"{exe_path}" "{pyc_file}"'
+    result = subprocess.run(command, capture_output=True, text=True, shell=True)
+
+    if result.returncode != 0:
+        raise Exception(f"Error running pycdc.exe: {result.stderr}")
+
+    return result.stdout
+
+
+def disassemble_pyc(file_path: str, pycdc_addr=None) -> str:
     """
     Disassembles a .pyc file using uncompyle6.
 
@@ -18,5 +45,11 @@ def disassemble_pyc(file_path: str) -> str:
         uncompyle6.main.decompile_file(file_path, output)
         return output.getvalue()
     except Exception as e:
-        print(f"Error occurred while disassembling: {e}")
+        if pycdc_addr is None:
-        return ""
+            print(
+                "ERROR: For Python 3.11 and above, you need to install pycdc and compile it yourself to obtain pycdc.exe."
+            )
+            print("repo: https://github.com/zrax/pycdc.git")
+            exit(1)
+        else:
+            return run_pycdc(pycdc_addr, file_path)

docs/design.md

@@ -1,9 +1,64 @@
 # 项目设计文档 - 后门检测系统
 
+## 打包
+
+### 简介
+
+本项目需要将 Python 代码打包成`pip`包和`deb`包，以便于分发和安装。以下是如何实现和使用该打包功能的详细步骤。
+
+### pip
+
+#### 打包命令
+
+```bash
+pip install wheel
+python setup.py sdist bdist_wheel
+```
+
+执行上述命令后，会在 dist 目录下生成 .tar.gz 和 .whl 文件。
+
+#### 本地安装
+
+- 安装 .whl 文件：
+
+``` bash
+pip install dist/backdoor_buster-0.1.0-py3-none-any.whl
+```
+
+- 安装 .tar.gz 文件：
+
+``` bash
+pip install dist/backdoor_buster-0.1.0.tar.gz
+```
+
+#### 上传到 PyPI
+
+- 安装 twine：
+
+``` bash
+pip install twine
+```
+
+- 使用 twine 上传包到 PyPI：
+
+``` bash
+twine upload dist/*
+```
+
+需要提供 PyPI 的用户名和密码。如果没有 PyPI 账号，可以在 PyPI 注册。
+
+#### 使用 PyPI 安装
+
+包上传到 PyPI 后，可以通过以下命令安装：
+
+``` bash
+pip install backdoor_buster
+```
+
 ## 静态代码后门检测
 
 **功能描述**:
-这个脚本用于扫描指定路径下的代码文件，检测潜在的危险函数调用，支持 `.py`, `.js`, `.cpp` 文件。
+这个脚本用于扫描指定路径下的代码文件，检测潜在的危险函数调用，支持 `.py`, `.js`, `.cpp`, `.pyc` 文件。
 
 **主要组件**:
 
@@ -67,7 +122,7 @@ python backdoor_detection.py ./src -o ./output/report.pdf
 **使用示例**:
 
 ```bash
-python requirements_detection.py ./requirements.txt -o ./output/report.md
+python -m detection.requirements_detection ./requirements.txt -o ./output/report.md
 ```
 
 ---

docs/tech_notes.md

@@ -46,7 +46,18 @@
 
 - **主要应用**：通过爬虫收集漏洞依赖信息并进行汇总，用于判断依赖是否存在漏洞版本。
 
-## 8. 代码和风险分析
+## 8. 打包
+
+本项目支持打包作为`pip`包进行发布
+
+- **主要应用**：
+  - `pip`通过`wheel`并自行撰写`setup.py`以及`MANIFEST.in`，将项目打包发布
+
+## 9. 反汇编
+
+项目通过`uncompyle6`库提供的反汇编模块可以实现对python字节码进行反汇编之后扫描危险代码
+
+## 10. 代码和风险分析
 
 项目中实现了基本的静态代码分析功能，用于识别和报告潜在的安全风险函数调用，如 `system`、`exec` 等。
 

docs/usage.md

@@ -2,31 +2,68 @@
 
 本文档提供了后门检测系统的使用方法，包括依赖版本漏洞检测和静态代码后门检测两部分。这将帮助用户正确执行安全检测，并理解输出结果。
 
-## 安装需求
-
-在开始使用本系统之前，请确保您的环境中安装了以下依赖：
-
-- Python 3.6 或更高版本
-- `packaging` 库：用于版本控制和比较
-- `reportlab` 库：用于生成 PDF 报告
-
-您可以通过以下命令安装必要的 Python 库：
-
-```bash
-pip install packaging reportlab
-```
-
 ## 下载和配置
 
 - 克隆或下载后门检测系统到您的本地环境。
 - 确保脚本文件 (`requirements_detection.py` 和 `backdoor_detection.py`) 在您的工作目录中。
 
+## 打包
+
+### pip
+
+#### 打包命令
+
+```bash
+pip install wheel
+python setup.py sdist bdist_wheel
+```
+
+执行上述命令后，会在 dist 目录下生成 .tar.gz 和 .whl 文件。
+
+#### 本地安装
+
+- 安装 .whl 文件：
+
+``` bash
+pip install dist/backdoor_buster-0.1.0-py3-none-any.whl
+```
+
+- 安装 .tar.gz 文件：
+
+``` bash
+pip install dist/backdoor_buster-0.1.0.tar.gz
+```
+
+#### 上传到 PyPI
+
+- 安装 twine：
+
+``` bash
+pip install twine
+```
+
+- 使用 twine 上传包到 PyPI：
+
+``` bash
+twine upload dist/*
+```
+
+需要提供 PyPI 的用户名和密码。如果没有 PyPI 账号，可以在 PyPI 注册。
+
+#### 使用 PyPI 安装
+
+包上传到 PyPI 后，可以通过以下命令安装：
+
+``` bash
+pip install backdoor_buster
+```
+
 ## 运行依赖版本漏洞检测脚本
 
 **命令格式**：
 
 ```bash
-python requirements_detection.py <requirements_file> -o <output_file> 
+python -m detection.requirements_detection <requirements_file> -o <output_file> 
 ```
 
 **参数说明**：
@@ -37,7 +74,7 @@ python requirements_detection.py <requirements_file> -o <output_file>
 **示例**：
 
 ```bash
-python requirements_detection.py requirements.txt -o output/report.md
+python -m detection.requirements_detection requirements.txt -o output/report.md
 ```
 
 ## 运行静态代码后门检测脚本
@@ -45,7 +82,7 @@ python requirements_detection.py requirements.txt -o output/report.md
 **命令格式**：
 
 ```bash
-python backdoor_detection.py <code_path> -o <output_file> -m <mode>
+python -m detection <code_path> -o <output_file> -m <mode>
 ```
 
 **参数说明**：
@@ -57,7 +94,7 @@ python backdoor_detection.py <code_path> -o <output_file> -m <mode>
 **示例**：
 
 ```bash
-python backdoor_detection.py ./src -o output/report.pdf -m regex
+python -m detection ./src -o output/report.pdf -m regex
 ```
 
 ## 结果解读

setup.py

@@ -0,0 +1,43 @@
+# pip install wheel
+# python setup.py sdist bdist_wheel
+
+from setuptools import setup, find_packages
+
+
+def read_file(filename: str) -> str:
+    """Read a file and return its content as a string.
+
+    Args:
+        filename (str): The name of the file to read.
+
+    Returns:
+        str: The content of the file.
+    """
+    with open(filename, encoding="utf-8") as f:
+        return f.read()
+
+
+setup(
+    name="backdoor_buster",
+    version="0.1.0",
+    author="ciscn",
+    description="A tool for integrated backdoor detection",
+    long_description=read_file("README.md"),
+    long_description_content_type="text/markdown",
+    url="https://git.mamahaha.work/sangge/BackDoorBuster",
+    packages=find_packages(),
+    classifiers=[
+        "Programming Language :: Python :: 3",
+        "License :: OSI Approved :: MIT License",
+        "Operating System :: OS Independent",
+    ],
+    python_requires=">=3.6",
+    install_requires=[
+        "reportlab",
+        "requests",
+        "packaging",
+        "openai",
+        "bs4",
+        "uncompyle6",
+    ],
+)

tests/test_backdoor_detection.py

@@ -1,7 +1,7 @@
 import unittest
 import warnings
 
-from detection.backdoor_detection import find_dangerous_functions
+from detection.__main__ import find_dangerous_functions
 from detection.GPTdetection import detectGPT
 import os