Cases › security breach › LR-2026-0239
BUGSTER uses DeepSeek-powered Hermes Agent to breach university MongoDB, stealing 5,200+ records
Whose AI: DeepSeek · DeepSeek V4security breach
Severity moderate: Other parties' data or systems were touched, and a source states how far it reached. Personal data of students and staff at two universities was stolen, a bounded scope of more than 5,200 records.
Earliest source we hold 2026-10-23
In one confirmed intrusion, the attacker stole a MongoDB database containing more than 5,200 records, including personal information about students and staff.
Risk grouping: AI used to exploit
| Who caused it | A person or organisation “A cybercriminal using the online handle BUGSTER used artificial intelligence tools to support attacks on Indian organizations, including two universities.” source |
|---|---|
| Intended or not | Intended “The actor configured prompt overrides and disabled local permission and sandbox controls, creating an environment intended to reduce restrictions on malicious tasks.” source |
| Before or after release | After release “A cybercriminal using the online handle BUGSTER used artificial intelligence tools to support attacks on Indian organizations, including two universities.” source |
| The AI's part | Used as a tool “Unit 42 said the attacker turned off safety controls and relied on AI-written scripts and reports.” source |
The story
1 · What it was doingBUGSTER ran the Hermes Agent framework with DeepSeek V4, plus Cursor and Telegram assistant profiles, to research targets, write scripts and reports, and scan for credentials against Indian universities.
2 · What went wrongThe AI-assisted tooling was used to break into university systems, including taking an administrator account protected by a weak password, crossing the universities' access boundaries.
3 · What it reached or didA MongoDB database with more than 5,200 student and staff records was stolen, and an e-commerce storefront API token, admin emails and payment merchant identifiers were collected.
4 · How it was caughtPalo Alto Networks' Unit 42 researchers identified and investigated the activity by examining the operator's environment, tools and records.
5 · What was doneThe source recommends reviewing administrator accounts, removing weak passwords, requiring multi-factor authentication, limiting database access and limiting AI agent permissions.
What the sources state
- Cause a permission, scope or wiring error
- What set it off misused by a person
Others like this
Show every field, with its source and grade
The AI system
Model
DeepSeek V4
C one secondary source states it
Developer
DeepSeek
C one secondary source states it
Software around the model
Hermes Agent
C one secondary source states it
Tools it was given
credential scanning, password cracking, coding, code generation (Cursor), Telegram bot assistant profiles, brute-force tooling, Wi-Fi capture and cloud GPU cracking, Discord webhooks
C one secondary source states it
What it was allowed to do
Attacker-configured environment with local permission and sandbox controls disabled, intended to reduce restrictions on malicious tasks.
C one secondary source states it
Stage of use
in live use
D inferred by us from the described behaviour, reason shown
Whose system
company using AI
D inferred by us from the described behaviour, reason shown
How it was connected
agentic with tools
D inferred by us from the described behaviour, reason shown
What it was for
AI assistant framework used to support attacks on Indian organizations, including mapping targets, writing scripts and reports
C one secondary source states it
State of its safeguards
safety controls turned off
C one secondary source states it
What happened
Earliest source we hold
2026-10-23
A the date the source itself carries
What set it off
misused by a person
C one secondary source states it
Who caught it
an outside party
D inferred by us from the described behaviour, reason shown
Themes
AI assisted attack, credential theft, database exfiltration, pii theft, education sector, jailbreak configuration
D inferred by us from the described behaviour, reason shown
Cause
a permission, scope or wiring error
D inferred by us from the described behaviour, reason shown
Cause in detail
The report attributes the intrusions to weak administrator account security and exposed services rather than a software flaw.
D inferred by us from the described behaviour, reason shown
Who caused it
an outside attacker
C one secondary source states it
What was done
Review administrator accounts, remove weak or reused passwords, require multi-factor authentication, limit database access, check internet-facing portals, monitor account activity, keep secrets out of public files and logs, revoke or rotate exposed credentials, and limit AI agent permissions with safety and approval controls enabled.
C one secondary source states it
What went wrong
security breach
C one secondary source states it
AI agent
an AI agent acted
C one secondary source states it
Impact
Severity
moderate
D inferred by us from the described behaviour, reason shown
Kind of harm
privacy, others' security
C one secondary source states it
Who was affected
two unnamed Indian universities; MongoDB database with student and faculty records
C one secondary source states it
Kinds of data exposed
personal information, credentials
C one secondary source states it
Why this severity
Other parties' data or systems were touched, and a source states how far it reached. Personal data of students and staff at two universities was stolen, a bounded scope of more than 5,200 records.
D inferred by us from the described behaviour, reason shown
Context
Industry
Educational Services
C one secondary source states it
Country
India
C one secondary source states it
LR-2026-0239 · version 2 · published 2026-10-25 after automatic checks; not yet checked field by field by a person · changes to this case · JSON