HackingIncidents.ai
Ledger / May 2026

Indonesian-speaking operator uses Claude Code skills to compromise government and business targets

Confirmed

CrowdStrike examined exposed offensive frameworks and documented successful AI-directed intrusions.

Model
Claude CodeAnthropic · closed
Agent harness
Claude Codevendor coding agent as operator
Attributed country
IDlanguage or infrastructure clues only
Sectors hit
government, enterprise broadIndonesian government domains, another country's government domain, and private-sector…

Observed during May–August 2026, the operator supplied Claude Code with 819 skills, primarily from public repositories. Human intervention addressed guardrails and stalled exploitation. The date represents the beginning of the reported observation window; the operation's exact start was not specified.

Sources

Date
May 2026
Disclosed
21 Oct 2026
Actor
Unnamed Indonesian-speaking operator (unknown), as attributed by CrowdStrike
Nexus
Indonesian-speaking; country attribution unestablished
Targets
Indonesian government domains, another country's government domain, and private-sector entities
What the AI did
Agentic intrusion
Model as named
Claude Code
Model basis
stated by researcher
Autonomy
semi-autonomous
Outcome
Multiple government and private-sector targets compromised.