Skip to content

ReD-Pot by RedSide Security ReD-Pot by RedSide Security

Internal-network deception: catch attackers after they get in.

Quick start How it works GitHub

ReD-Pot Community Edition

ReD-Pot places decoy servers inside your network and makes them look like your own: a Windows file server, an Active Directory domain controller, a Linux web server, a backup server or a DNS server, all carrying your domain, your host-naming scheme and believable users and files. Real users have no reason to touch them, so any connection to a decoy is a signal, and a stolen bait credential being used is close to proof of an intrusion.

This site documents ReD-Pot Community Edition (CE) 1.0.1, the free, open-source (MIT) edition published at github.com/RedSideSecurity/ReD-Pot. Features that exist only in ReD-Pot Enterprise are marked with an Enterprise feature box.

What ReD-Pot does

  • Decoy servers


    Five roles: Windows file server, Windows domain controller (LDAP, Kerberos, DNS, SYSVOL), Linux web server, FTP/backup server and Linux DNS server. Each decoy answers on the IP of its own network interface.

    Decoy roles

  • Baits and honeytokens


    Planted passwords, SSH keys, .env files, database dumps, a GPP password, a Kerberoastable account and Word documents that report when opened. Each one has a unique id.

    Honeytokens

  • Severity ladder L1 to L5


    Events are grouped into incidents per source and scored by how far the source got, from a TCP touch to a stolen credential being used.

    Incidents and levels

  • Alerts for your SIEM


    Incidents at L3 and above are written to alerts.json (JSON Lines) for Wazuh, Splunk, ELK or any log shipper, and can be exported as one JSON file.

    Alerts, SIEM and export

Where it fits in an intrusion

Most intrusions do not stop at the first compromised machine. After initial access, the attacker has to look around, collect credentials and move to other servers. ReD-Pot sits in that window:

Stage What the attacker does What ReD-Pot sees
Discovery port scans, share listings, LDAP queries, DNS zone transfers L3 intent: someone is mapping a server nobody uses
Credential access reads passwords.txt, .env, GPP passwords, Kerberoasts a service account L4 compromise: a bait file was read
Lateral movement uses the stolen password or key on the next server L5 honeytoken: a bait credential was used, with the workstation and user

An L5 alert reads like this: "password planted on RS-FS-01 used on RS-WEB-01 by ACC-PC-07 (anna.k)". The attacker gains nothing useful, because every file and password on a decoy is fake.

Architecture at a glance

flowchart LR
    subgraph LAN["Your LAN / VLANs"]
        WS["Workstation<br/>(attacker foothold)"]
    end
    subgraph MGR["Manager server"]
        D1["Decoy: file server<br/>own interface + IP"]
        D2["Decoy: domain controller<br/>own interface + IP"]
        C["Manager console<br/>web UI + sensor API<br/>(management interface only)"]
    end
    subgraph REM["Remote site"]
        S["Remote sensor<br/>decoy + relay"]
    end
    WS -- "scan, logon, steal bait" --> D1
    WS -- "reuse stolen credential" --> D2
    WS -.-> S
    D1 -- events --> C
    D2 -- events --> C
    S -- "pinned TLS" --> C
    C --> J["alerts.json<br/>to your SIEM"]

See Architecture for the components and how they talk to each other.

Community Edition and Enterprise

Community Edition Enterprise
Licence MIT, free commercial licence
Decoy roles, baits, honeytokens, L1-L5 scoring, network map, world map, wall screen
alerts.json / events.json for your SIEM, JSON export
Remote sensors with pinned TLS
In-place update (install.sh update) signed offline bundles
Number of decoys up to 3 more than 3 (set by the licence)
Decoy placement one decoy per interface, on that interface's own IP also extra IPs, VLANs and vendor MAC addresses
Outbound alerts (Telegram, e-mail, webhook) and alert rules -
Sensor offline spool and replay -
Block an attacker IP on all decoys, country filter -
Editable persona and users -
Retention per data type, custom TLS certificate, audit log -

The full comparison is on Community and Enterprise.

Next steps

Authorised use only

Use ReD-Pot only on networks you own or are authorised to protect.