Update README.md

Added more info about the `details` and `interactive` fields to logging section.

README.md

@@ -1,5 +1,7 @@
 # DECEIVE
 
+<img align="right" src="DECEIVE.png" alt="A cybercriminal interacts with a ghostly, AI-driven honeypot system">
+
 DECEIVE, the **DECeption with Evaluative Integrated Validation Engine**, is a high-interaction, low-effort honeypot system. Unlike most high-interaction honeypots, DECEIVE doesn't provide attackers with access to any actual system. AI actually does all the work of simulating a realistic honeypot system based on a configurable system prompt that describes what type of system you want to simulate. Unlike many other high-interaction honeypots which require substantial effort to seed with realistic users, data, and applications, DECEIVE's AI backend will do all this for you, automatically.
 
 This version of DECEIVE simulates a Linux server via the SSH protocol. It will log all the user inputs, the outputs returned by the LLM backend, as well as a summary of each session after they end. It'll even tell you if it thinks a users' session was benign, suspicious, or outright malicious.
@@ -91,8 +93,8 @@ Things to note:
     * `Session summary`
     * `SSH connection closed`
 * Several of these message types also feature a `details` field with additional information
-    * `User input` messages contain a base64-encoded copy of the entire user input
+    * `User input` messages contain a base64-encoded copy of the entire user input in the `details` field, as well as an `interactive` field (true/false) that tells you whether this was an interactive or non-interactive command (i.e., whether they logged in with a terminal session or provided a command on the SSH command-line).
-    * `LLM response` messages contain a base64-encoded copy of the entire simulated response
+    * `LLM response` messages contain a base64-encoded copy of the entire simulated response in the `details` field.
     * `Session summary` messages contain not only a summary of the commands, but also a guess as to what they might have been intended to accomplish. There will also be a `judgement` field that contains one of "BENIGN", "SUSPICIOUS", or "MALICIOUS"
 * Since this is a honeypot and not intended for use by real users, IT WILL LOG USERNAMES AND PASSWORDS! These are found in the `Authentication success` messages, in the `username` and `password` fields.
 

SSH/config.ini.TEMPLATE

@@ -51,10 +51,11 @@ system_prompt = Interpret all inputs as though they were SSH commands and provid
     responses do not indicate that the user is on a virtual machine, container, 
     or emulation at all. 
 
-    On the first call, be sure to include a realistic MOTD.
+    If {interactive} is True, make sure to include a realistic MOTD on the
+    first call, and to end all responses with a realistic shell prompt to display to 
+    the user, including a space at the end. 
 
-    End all responses with a realistic shell prompt to display to the user, 
+    If {interactive} is False, do not include a shell prompt or MOTD.
-    including a space at the end.
 
     Include ANSI color codes for the terminal with the output of ls commands 
     (including any flags), or in any other situation where it is appropriate, but 
@@ -74,9 +75,12 @@ system_prompt = Interpret all inputs as though they were SSH commands and provid
 # The valid user accounts and passwords for the SSH server, in the
 # form "username = password".  Note that you can enable login without 
 # a password by leaving that field blank (e.g., "guest =" on a line by
-# itself)
+# itself). You can set an account to accept ANY password, including an empty
+# password, by setting the password to "*"
 [user_accounts]
 guest = 
 user1 = secretpw
 user2 = password123
+root = *
+
 

SSH/ssh_server.py

@@ -1,11 +1,14 @@
 #!/usr/bin/env python3
 
 from configparser import ConfigParser
+import argparse
 import asyncio
 import asyncssh
 import threading
 import sys
 import json
+import os
+import traceback
 from typing import Optional
 import logging
 import datetime
@@ -20,6 +23,7 @@ from langchain_core.chat_history import BaseChatMessageHistory, InMemoryChatMess
 from langchain_core.runnables.history import RunnableWithMessageHistory
 from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
 from langchain_core.runnables import RunnablePassthrough
+from asyncssh.misc import ConnectionLost
 
 class JSONFormatter(logging.Formatter):
     def format(self, record):
@@ -33,6 +37,8 @@ class JSONFormatter(logging.Formatter):
             "dst_port": record.dst_port,
             "message": record.getMessage()
         }
+        if hasattr(record, 'interactive'):
+            log_record["interactive"] = record.interactive
         # Include any additional fields from the extra dictionary
         for key, value in record.__dict__.items():
             if key not in log_record and key != 'args' and key != 'msg':
@@ -46,8 +52,18 @@ class MySSHServer(asyncssh.SSHServer):
 
     def connection_made(self, conn: asyncssh.SSHServerConnection) -> None:
         # Get the source and destination IPs and ports
-        (src_ip, src_port, _, _) = conn.get_extra_info('peername')
+        peername = conn.get_extra_info('peername')
-        (dst_ip, dst_port, _, _) = conn.get_extra_info('sockname')
+        sockname = conn.get_extra_info('sockname')
+
+        if peername is not None:
+            src_ip, src_port = peername[:2]
+        else:
+            src_ip, src_port = '-', '-'
+
+        if sockname is not None:
+            dst_ip, dst_port = sockname[:2]
+        else:
+            dst_ip, dst_port = '-', '-'
 
         # Store the connection details in thread-local storage
         thread_local.src_ip = src_ip
@@ -61,6 +77,8 @@ class MySSHServer(asyncssh.SSHServer):
     def connection_lost(self, exc: Optional[Exception]) -> None:
         if exc:
             logger.error('SSH connection error', extra={"error": str(exc)})
+            if not isinstance(exc, ConnectionLost):
+                traceback.print_exception(exc)
         else:
             logger.info("SSH connection closed")
         # Ensure session summary is called on connection loss if attributes are set
@@ -87,7 +105,7 @@ class MySSHServer(asyncssh.SSHServer):
     def validate_password(self, username: str, password: str) -> bool:
         pw = accounts.get(username, '*')
         
-        if ((pw != '*') and (password == pw)):
+        if pw == '*' or (pw != '*' and password == pw):
             logger.info("Authentication success", extra={"username": username, "password": password})
             return True
         else:
@@ -134,7 +152,8 @@ representative examples.
     llm_response = await session.ainvoke(
         {
             "messages": [HumanMessage(content=prompt)],
-            "username": process.get_extra_info('username')
+            "username": process.get_extra_info('username'),
+            "interactive": True  # Ensure interactive flag is passed
         },
             config=llm_config
     )
@@ -162,32 +181,47 @@ async def handle_client(process: asyncssh.SSHServerProcess, server: MySSHServer)
 
     llm_config = {"configurable": {"session_id": task_uuid}}
 
+    try:
+        if process.command:
+            # Handle non-interactive command execution
+            command = process.command
+            logger.info("User input", extra={"details": b64encode(command.encode('utf-8')).decode('utf-8'), "interactive": False})
+            llm_response = await with_message_history.ainvoke(
+                {
+                    "messages": [HumanMessage(content=command)],
+                    "username": process.get_extra_info('username'),
+                    "interactive": False
+                },
+                    config=llm_config
+            )
+            process.stdout.write(f"{llm_response.content}")
+            logger.info("LLM response", extra={"details": b64encode(llm_response.content.encode('utf-8')).decode('utf-8'), "interactive": False})
+            await session_summary(process, llm_config, with_message_history, server)
+            process.exit(0)
+        else:
+            # Handle interactive session
             llm_response = await with_message_history.ainvoke(
                 {
                     "messages": [HumanMessage(content="ignore this message")],
-            "username": process.get_extra_info('username')
+                    "username": process.get_extra_info('username'),
+                    "interactive": True
                 },
                     config=llm_config
             )
 
             process.stdout.write(f"{llm_response.content}")
-    logger.info("LLM response", extra={"details": b64encode(llm_response.content.encode('utf-8')).decode('utf-8')})
+            logger.info("LLM response", extra={"details": b64encode(llm_response.content.encode('utf-8')).decode('utf-8'), "interactive": True})
 
-    # Store process, llm_config, and session in the MySSHServer instance
-    server._process = process
-    server._llm_config = llm_config
-    server._session = with_message_history
-
-    try:
             async for line in process.stdin:
                 line = line.rstrip('\n')
-            logger.info("User input", extra={"details": b64encode(line.encode('utf-8')).decode('utf-8')})
+                logger.info("User input", extra={"details": b64encode(line.encode('utf-8')).decode('utf-8'), "interactive": True})
 
                 # Send the command to the LLM and give the response to the user
                 llm_response = await with_message_history.ainvoke(
                     {
                         "messages": [HumanMessage(content=line)],
-                    "username": process.get_extra_info('username')
+                        "username": process.get_extra_info('username'),
+                        "interactive": True
                     },
                         config=llm_config
                 )
@@ -197,7 +231,7 @@ async def handle_client(process: asyncssh.SSHServerProcess, server: MySSHServer)
                     return
                 else:
                     process.stdout.write(f"{llm_response.content}")
-                logger.info("LLM response", extra={"details": b64encode(llm_response.content.encode('utf-8')).decode('utf-8')})
+                    logger.info("LLM response", extra={"details": b64encode(llm_response.content.encode('utf-8')).decode('utf-8'), "interactive": True})
 
     except asyncssh.BreakReceived:
         pass
@@ -285,10 +319,24 @@ def choose_llm():
 
     return llm_model
 
-def get_prompts() -> dict:
+def get_prompts(prompt: Optional[str], prompt_file: Optional[str]) -> dict:
     system_prompt = config['llm']['system_prompt']
+    if prompt is not None:
+        if not prompt.strip():
+            print("Error: The prompt text cannot be empty.", file=sys.stderr)
+            sys.exit(1)
+        user_prompt = prompt
+    elif prompt_file:
+        if not os.path.exists(prompt_file):
+            print(f"Error: The specified prompt file '{prompt_file}' does not exist.", file=sys.stderr)
+            sys.exit(1)
+        with open(prompt_file, "r") as f:
+            user_prompt = f.read()
+    elif os.path.exists("prompt.txt"):
         with open("prompt.txt", "r") as f:
             user_prompt = f.read()
+    else:
+        raise ValueError("Either prompt or prompt_file must be provided.")
     return {
         "system_prompt": system_prompt,
         "user_prompt": user_prompt
@@ -296,12 +344,25 @@ def get_prompts() -> dict:
 
 #### MAIN ####
 
+try:
+    # Parse command line arguments
+    parser = argparse.ArgumentParser(description='Start the SSH honeypot server.')
+    parser.add_argument('-c', '--config', type=str, default='config.ini', help='Path to the configuration file')
+    parser.add_argument('-p', '--prompt', type=str, help='The entire text of the prompt')
+    parser.add_argument('-f', '--prompt-file', type=str, default='prompt.txt', help='Path to the prompt file')
+    args = parser.parse_args()
+
+    # Check if the config file exists
+    if not os.path.exists(args.config):
+        print(f"Error: The specified config file '{args.config}' does not exist.", file=sys.stderr)
+        sys.exit(1)
+
     # Always use UTC for logging
     logging.Formatter.formatTime = (lambda self, record, datefmt=None: datetime.datetime.fromtimestamp(record.created, datetime.timezone.utc).isoformat(sep="T",timespec="milliseconds"))
 
     # Read our configuration file
     config = ConfigParser()
-config.read("config.ini")
+    config.read(args.config)
 
     # Read the user accounts from the configuration file
     accounts = get_user_accounts()
@@ -320,7 +381,7 @@ logger.addFilter(f)
 
     # Now get access to the LLM
 
-prompts = get_prompts()
+    prompts = get_prompts(args.prompt, args.prompt_file)
     llm_system_prompt = prompts["system_prompt"]
     llm_user_prompt = prompts["user_prompt"]
 
@@ -371,3 +432,8 @@ asyncio.set_event_loop(loop)
     loop.run_until_complete(start_server())
     loop.run_forever()
 
+except Exception as e:
+    print(f"Error: {e}", file=sys.stderr)
+    traceback.print_exc()
+    sys.exit(1)
+