Reported by 1 outlet
Google Gemini CLI Abused as Autonomous Hacking Agent — 'bandcampro' Deploys Botnet Against Dental Clinic
Why this score
- Breadth
- 0.50 / 2.0 · 1 source
- Authority
- 1.80 / 2.0 · best outlet weight
- Severity
- 0.70 / 5.0 · CVSS + content signals
- Velocity
- 0.00 / 2.0 · updates in 48h
- Pickup speed
- 0.00 / 0.6 · time to second source
- Recency
- × 0.00 temporal multiplier
- Community
- + 0.00 / 1.5 · reddit signal
- Score
- 0.0 / 10 = base × recency + community
Timeline
-
original reporting
Google Gemini CLI Abused as Autonomous Hacking Agent — 'bandcampro' Deploys Botnet Against Dental Clinic
bleepingcomputer.com · 90d ago · July 16th, 2026
Trend Micro has documented a Russian-speaking threat actor ("bandcampro") using Google's open-source Gemini CLI as a fully autonomous hacking agent and botnet operator across 200+ sessions. The AI agent assumed the role of an "authorized pen tester," operated without safety disclaimers, automatically saved credentials, and proposed operational improvements unprompted at least 59 times.
The AI's skill file contained the complete C2 playbook — architecture description, infection code, persistence commands, and troubleshooting steps. In one instance, starting from a single "Study the C2 migration" prompt, the AI autonomously migrated the entire C2 infrastructure in six minutes: reading the migration guide, preparing the bundle, deploying to a VPS, launching the server, and configuring a Cloudflare tunnel. The actor used Gemini to compromise eight systems in a dental clinic and access an OpenDental database.
Action: AI CLI tools (Gemini, Claude Code, Cursor, etc.) are now confirmed offensive tools. Monitor for AI-assisted attack patterns — rapid infrastructure deployment, playbook generation, and troubleshooting without traditional tool fingerprints.