Lucene search
+L

BloodHound.py collect

🗓️ 10 Aug 2026 06:19:08Reported by KittySploitType 
kittysploit
 kittysploit
🔗 kittysploit.com

BloodHound.py collect runs bloodhound-python against a Domain Controller to produce a zip.

Refs
Code
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""Collect AD attack-path data via bloodhound-python (Linux/Windows, no SharpHound.exe)."""

from kittysploit import *
from lib.protocols.ldap.bloodhound_python_collect import (
    bloodhound_python_available,
    collect_and_merge_kb,
    run_bloodhound_python,
)


class Module(Scanner):
    __info__ = {
        "name": "BloodHound.py collect",
        "description": (
            "Run bloodhound-python / bloodhound-ce-python against a Domain Controller "
            "from Linux or Windows, produce a zip, and merge into the agent attack graph. "
            "Default collection=DCOnly (LDAP-only, good from Kali). Use All for fuller "
            "Session/LocalAdmin fan-out when the network allows."
        ),
        "author": ["KittySploit Team"],
        "severity": "info",
        "modules": [],
        "tags": ["ad", "ldap", "bloodhound", "graph", "scanner"],
        "references": [
            "https://github.com/dirkjanm/BloodHound.py",
            "https://www.kali.org/tools/bloodhound.py/",
        ],
        "agent": {
            "risk": "intrusive",
            "effects": ["active_exploitation"],
            "expected_requests": 20,
            "reversible": False,
            "approval_required": True,
            "produces": ["risk_signals"],
            "cost": 1.6,
            "noise": 0.55,
            "value": 0.9,
            "requires": {
                "capabilities_any": [],
                "capabilities_all": [],
            },
            "chain": {
                "consumes_capabilities": [],
                "produces_capabilities": ["ad_graph"],
            },
        },
    }

    domain = OptString("", "AD domain FQDN (e.g. CORP.LOCAL)", True)
    username = OptString("", "Domain username", True)
    password = OptString("", "Password (or leave empty and set hashes)", False)
    hashes = OptString("", "LM:NT or :NT hash (Impacket style)", False)
    nameserver = OptString("", "DNS / DC IP for -ns (recommended)", False)
    domain_controller = OptString("", "Override DC hostname (-dc)", False)
    collection = OptString(
        "DCOnly",
        "Collection methods: DCOnly, Default, All, Group, ACL, …",
        False,
    )
    use_ldaps = OptBool(False, "Use LDAPS", False)
    timeout = OptInteger(900, "Collector timeout seconds", False)
    import_graph = OptBool(True, "Merge into agent attack_graph after collect", False)
    replace = OptBool(True, "Replace existing attack_graph on import", False)

    def run(self):
        avail = bloodhound_python_available()
        if not avail.get("available"):
            print_error("bloodhound-python not installed")
            print_info(avail.get("install_hint") or "pip install bloodhound")
            return False

        domain = str(self.domain or "").strip()
        username = str(self.username or "").strip()
        password = str(self.password or "")
        hashes = str(self.hashes or "").strip()
        if not domain or not username:
            print_error("domain and username are required")
            return False
        if not password and not hashes:
            print_error("password or hashes required")
            return False

        kwargs = dict(
            domain=domain,
            username=username,
            password=password,
            hashes=hashes,
            nameserver=str(self.nameserver or "").strip(),
            domain_controller=str(self.domain_controller or "").strip(),
            collection=str(self.collection or "DCOnly"),
            use_ldaps=bool(self.use_ldaps),
            timeout_sec=int(self.timeout or 900),
            replace=bool(self.replace),
        )

        print_status(
            f"bloodhound-python collect domain={domain} user={username} "
            f"collection={kwargs['collection']} cli={avail.get('cli')}"
        )

        if self.import_graph:
            result = collect_and_merge_kb(getattr(self, "framework", None), **kwargs)
        else:
            result = run_bloodhound_python(**{k: v for k, v in kwargs.items() if k != "replace"})

        if not result.get("success"):
            print_error(result.get("error") or "Collection failed")
            if result.get("stderr"):
                print_info(str(result["stderr"])[:2000])
            if result.get("install_hint"):
                print_info(result["install_hint"])
            return False

        export_path = result.get("export_path") or result.get("zip_path")
        print_success(f"BloodHound export: {export_path}")
        if result.get("imported"):
            print_success(f"Merged into attack_graph (+{result.get('nodes_added', 0)} nodes)")
        print_info(
            "Import into Cosmic Attack Graph: Attack Graph → file import, "
            "or POST /api/attack-graph/import with this zip path."
        )
        self.vulnerability_info = {
            "export_path": export_path,
            "collection": result.get("collection"),
            "cli": result.get("cli"),
            "imported": bool(result.get("imported")),
        }
        return True

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

10 Aug 2026 06:19Current
5.9Medium risk
Vulners AI Score5.9
0