Aoki infostealer targets Claude Code, ChatGPT, Codex, Cursor tokens
Aoki (aka Aokigahara) infostealer added AI credential theft in 2026. Targets: .claude.json, Claude desktop cookies, ChatGPT cookies, Codex config.toml/.env (github.com/openai/codex-cli), GitHub Copilot…
What happened
Aoki (aka Aokigahara) infostealer added AI credential theft in 2026. Targets: .claude.json, Claude desktop cookies, ChatGPT cookies, Codex config.toml/.env (github.com/openai/codex-cli), GitHub Copilot tokens (github.com/features/copilot), Cursor state.vscdb/storage.json (github.com/getcursor/cursor), Gemini CLI, JetBrains AI Assistant, Kiro, Windsurf, Codeium (github.com/Exafunction/codeium.vim), AWS credentials/SSO, Amazon Q, Git credentials, SSH keys. Cursor local state caches chat/code. Stolen keys allow acting under developer identity/quota, expose repos/cloud/CI-CD, bypass safeguards. Capable of LLMjacking. Earlier 2026 versions didn't target AI; capability recently added/expanding. Detection: monitor AI credential file access (.claude.json, .codex/auth.json, Cursor globalStorage), C2 domains (dobeqnesmot.cc, hjazfioe.surf), API usage anomalies from unexpected geolocations.
Recent Aoki/Aokigahara infostealer builds steal AI coding assistant credentials, session files, and tokens
How it happened
Read the flow as text & sources
A catalog reconstruction of the supplied source account. Unknown steps remain explicit; an edge describes the reported relationship between steps.
- 01 · Deploy infostealer
Aoki/Aokigahara deployed (distribution method not disclosed)
Source for step 1 - 02 · AI credential harvesting
Steals .claude.json, Claude/ChatGPT cookies, Codex config, Copilot/Cursor/Gemini/JetBrains/Kiro/Windsurf/Codeium tokens, AWS creds, Git creds, SSH keys; Cursor state caches chat/code
Source for step 2 - 03 · Traditional targets
Browsers, crypto wallets, password managers, messaging/gaming, VPN/FTP, email
Source for step 3 - 04 · C2 exfiltration
Sends data to C2 (dobeqnesmot.cc, hjazfioe.surf); no backdoor commands
Source for step 4 - 05 · Credential abuse
Stolen keys enable acting as developer, LLMjacking, accessing repos/cloud/CI-CD, resale
Source for step 5
Why it matters
Reported entry point: Infostealer distribution (method not disclosed); likely traditional vectors. Reported impact: Sends data to C2 (dobeqnesmot.cc, hjazfioe.surf); no backdoor commands; Stolen keys enable acting as developer, LLMjacking, accessing repos/cloud/CI-CD, resale; reported targets include Developers using AI coding assistants; scale not disclosed; applicability to your setup has not been assessed, so verify affected versions, permissions, and configuration in the original report.
What you can do
Protection & prevention
Rotate AI API keys after compromise; scope keys narrowly; monitor unusual API usage; endpoint protection; credential managers; monitor Aoki C2 domains; file integrity monitoring for AI credential files
Catalog defensive guidance based on the reported behavior. Confirm scope and current mitigations in the original source.
Technical detailsTooling, impact & MITRE ATLAS
Attack context
- AI's role
- AI credential theft
- Initial access
- Infostealer distribution (method not disclosed); likely traditional vectors
Tooling as reported
AI products, models and artifacts
- Claude Code
- Claude Desktop
- ChatGPT Desktop
- OpenAI Codex
- Cursor
- GitHub Copilot
- Google Gemini CLI
- JetBrains AI Assistant
- Kiro IDE
- Windsurf
- Codeium
- Amazon Q
Other tooling
- Aoki/Aokigahara infostealer
- AWS CLI/SSO
- Git
- SSH
Targets and reported impact
- Targets / sectors
- Developers using AI coding assistants; scale not disclosed
- Reported scale / exposure
- Capability added 2026, expanding; no victim count
Reported attribution
Unknown; no links to existing families; new stealer first seen earlier 2026
Source defensive notes
Rotate AI API keys after compromise; scope keys narrowly; monitor unusual API usage; endpoint protection; credential managers; monitor Aoki C2 domains; file integrity monitoring for AI credential files
MITRE ATLAS
AI-specific catalog annotations based on the reported behavior; these are not MITRE case-study assignments.
Credential theft targeting AI development tools and cloud credentials Source
Indicators of compromise
domains
dobeqnesmot.cchjazfioe.surf
ips
hashes
urls
Sources & research credit
3 sourcesResearch by Symantec Threat Hunter Team, Check Point Research · Broadcom (Symantec), Check Point
- Aoki: Infostealer Targets AI Coding Assistant Credentials www.broadcom.com
- Symantec Aoki AI targeting www.security.com
- Check Point: Infostealer riding the AI Train research.checkpoint.com
OS not specified; other targets suggest Windows; macOS/Linux not confirmed. Detection guidance: monitor AI credential file access, C2 communication (dobeqnesmot.cc, hjazfioe.surf), API usage anomalies from unexpected locations, unauthorized reads of .claude.json, .codex/auth.json, Cursor globalStorage, .copilot/config.json, .gemini/config.json, .aws/credentials, SSH keys.
Record & downloads
- Source record name
- Aoki infostealer targets Claude Code, ChatGPT, Codex, Cursor tokens
- Entry ID
- symantec-aoki-infostealer-ai-2026-10
- Catalog review
- November 4, 2026