📜 Honey GPO
A decoy Group Policy object, and a SYSVOL preference file, that nothing legitimate ever reads.
What it is
Create a Group Policy object linked to an empty OU, and place a Group Policy Preferences file in its SYSVOL folder holding a decoy cpassword. No computer or user is in scope, so nothing legitimate reads either. GPO enumeration (Get-DomainGPO, BloodHound) reads the decoy GPO object, and GPP password harvesting (Get-GPPPassword) reads the decoy file. Both reads are audited.
Implementation requirements
- A GPO linked to an empty or isolated OU, named in the estate's own GPO naming convention
- Audit Directory Service Access enabled, with a SACL on the decoy groupPolicyContainer object so reads raise event 4662
- A Groups.xml preference file in the decoy GPO's SYSVOL folder holding a decoy cpassword for a disabled, privilege-free account; audit access to it (event 4663), or deny Everyone and audit the failures, as Metcalf (2015) suggests
- Alert on any 4662 or 4663 against the decoy objects, and on any logon attempt with the decoy account
- Safety: the decoy account is disabled and holds no rights; the GPO applies to no real machine
- Rollback: Unlink and delete the GPO and its SYSVOL files, then disable the decoy account.
- Upkeep: Check after domain changes that the OU stays empty and the SACL remains. Quarterly proof-of-life read.
Cost
Negligible. A GPO and a SYSVOL file are free; audit volume is low because nothing legitimate touches them.
Indicative only. Cloud pricing and licensing change — check the provider's official pricing pages for current rates before you budget.
Technologies
ATT&CK coverage 2
| Technique | Tactic | Status | Confidence | Tier |
|---|---|---|---|---|
| T1552.006 Group Policy Preferences |
Credential Access | confirmed | strong | T2 |
| ↳ Detection point: the attack step itself touches the decoy | ||||
↳ Evidence (3 sources)
| ||||
| T1615 Group Policy Discovery |
Discovery | theoretical | moderate | T2 |
| ↳ Detection point: the attack step itself touches the decoy | ||||
↳ Evidence (3 sources)
Moderate: the behaviour and the audit signal are documented; no study or product shows a decoy GPO caught this way. | ||||
Evidence
- Decoys Cannot Go Everywhere: Mapping the Deception Surface in MITRE ATT&CK (companion data) — Valeros, V.; Catania, C.; Lisý, V.; Griffioen, H., Stratosphere Laboratory, Czech Technical University in Prague 2026 preprint
- Finding Passwords in SYSVOL & Exploiting Group Policy Preferences — Metcalf, S., ADSecurity.org 2015 practitioner
- 4662(S, F) An operation was performed on an object — Microsoft industry
- 4663(S) An attempt was made to access an object — Microsoft industry
- MITRE ATT&CK Enterprise v20.0 — MITRE 2026 standard
Research basis: Metcalf (ADSecurity.org, 2015) recommends a decoy GPP XML file in SYSVOL with access auditing, since no legitimate GPO references it. GPP password harvesting is documented attacker tradecraft (ATT&CK T1552.006: Wizard Spider, APT33). Valeros et al. (2026) rate T1615 Group Policy Discovery as highly feasible with a honey GPO.