⤷ Title: Chat History Poisoning in LLM Applications: Persistent Prompt Injection via Untrusted JSON Context
════════════════════════
𐀪 Author: Serhat ÇİÇEK
════════════════════════
ⴵ Time: Thu, 27 Nov 2025 14:11:54 GMT
════════════════════════
⌗ Tags: #context_engineering #agentic_ai #prompt_injection_attack #application_security #llm_applications
════════════════════════
𐀪 Author: Serhat ÇİÇEK
════════════════════════
ⴵ Time: Thu, 27 Nov 2025 14:11:54 GMT
════════════════════════
⌗ Tags: #context_engineering #agentic_ai #prompt_injection_attack #application_security #llm_applications
Medium
Chat History Poisoning in LLM Applications: Persistent Prompt Injection via Untrusted JSON Context
LLM security guide: How chat history manipulation can be exploited and mitigated
⤷ Title: AI Clutter Solved? Windows 11 Adds Setting to Disable AI Actions in File Explorer Menu
════════════════════════
𐀪 Author: Ddos
════════════════════════
ⴵ Time: Tue, 09 Dec 2025 00:03:54 +0000
════════════════════════
⌗ Tags: #Windows #AI Actions #Context Menu #File Explorer #Microsoft Copilot #UI Clutter #User Feedback #Windows 11 #Windows Build 26220
════════════════════════
𐀪 Author: Ddos
════════════════════════
ⴵ Time: Tue, 09 Dec 2025 00:03:54 +0000
════════════════════════
⌗ Tags: #Windows #AI Actions #Context Menu #File Explorer #Microsoft Copilot #UI Clutter #User Feedback #Windows 11 #Windows Build 26220
Daily CyberSecurity
AI Clutter Solved? Windows 11 Adds Setting to Disable AI Actions in File Explorer Menu
Responding to user complaints, Windows 11 Build 26220 adds a setting to disable AI actions in the File Explorer context menu, reducing significant UI clutter.
⤷ Title: Breaking the AI Silo: How Google’s New “Import AI Chats” Feature Ends Context Lock-In
════════════════════════
𐀪 Author: Ddos
════════════════════════
ⴵ Time: Wed, 04 Feb 2026 03:00:35 +0000
════════════════════════
⌗ Tags: #Technology #AI chat history #AI data portability #ChatGPT #Claude #context lock_in #Data Migration #Gemini Beta features #Google Gemini #Import AI Chats #Tech News 2026
════════════════════════
𐀪 Author: Ddos
════════════════════════
ⴵ Time: Wed, 04 Feb 2026 03:00:35 +0000
════════════════════════
⌗ Tags: #Technology #AI chat history #AI data portability #ChatGPT #Claude #context lock_in #Data Migration #Gemini Beta features #Google Gemini #Import AI Chats #Tech News 2026
Daily CyberSecurity
Breaking the AI Silo: How Google’s New "Import AI Chats" Feature Ends Context Lock-In
Google Gemini is testing an "Import AI Chats" tool to let users migrate histories from ChatGPT and Claude. Say goodbye to starting over with every new AI.
⤷ Title: The curious case of poisoned context
════════════════════════
𐀪 Author: Elina Maliarsky
════════════════════════
ⴵ Time: Wed, 04 Mar 2026 09:35:28 GMT
════════════════════════
⌗ Tags: #ethical_hacking #agentic_ai #context_engineering #generative_ai_tools #large_language_models
════════════════════════
𐀪 Author: Elina Maliarsky
════════════════════════
ⴵ Time: Wed, 04 Mar 2026 09:35:28 GMT
════════════════════════
⌗ Tags: #ethical_hacking #agentic_ai #context_engineering #generative_ai_tools #large_language_models
Medium
The curious case of poisoned context
A Red Team Mystery in Six Acts
⤷ Title: OpenAI Unveils GPT-5.4: The “Professional” Frontier with Native Computer Control and 1M Token Context
════════════════════════
𐀪 Author: Ddos
════════════════════════
ⴵ Time: Fri, 06 Mar 2026 02:19:50 +0000
════════════════════════
⌗ Tags: #Technology #AI Agents #API pricing #Computer_Use AI #Context Window #GPT_5.4 #GPT_5.4 Pro #OpenAI #OSWorld #presentation generation #spreadsheet modeling #Tech News 2026
════════════════════════
𐀪 Author: Ddos
════════════════════════
ⴵ Time: Fri, 06 Mar 2026 02:19:50 +0000
════════════════════════
⌗ Tags: #Technology #AI Agents #API pricing #Computer_Use AI #Context Window #GPT_5.4 #GPT_5.4 Pro #OpenAI #OSWorld #presentation generation #spreadsheet modeling #Tech News 2026
Daily CyberSecurity
OpenAI Unveils GPT-5.4: The "Professional" Frontier with Native Computer Control and 1M Token Context
OpenAI launches GPT-5.4, featuring native computer control that beats human benchmarks and a 1M token context window. See the new "Professional" AI in action.
⤷ Title: Breaking the App Shell: Five New Electron Vulnerabilities Shatter Context Isolation
════════════════════════
𐀪 Author: Ddos
════════════════════════
ⴵ Time: Tue, 07 Apr 2026 12:07:02 +0000
════════════════════════
⌗ Tags: #Vulnerability Report #Context Isolation #CVE_2026_34780 #cybersecurity #Electron #infosec #memory corruption #Node.js #Sandbox Escape #use after free #web development
════════════════════════
𐀪 Author: Ddos
════════════════════════
ⴵ Time: Tue, 07 Apr 2026 12:07:02 +0000
════════════════════════
⌗ Tags: #Vulnerability Report #Context Isolation #CVE_2026_34780 #cybersecurity #Electron #infosec #memory corruption #Node.js #Sandbox Escape #use after free #web development
Daily CyberSecurity
Breaking the App Shell: Five New Electron Vulnerabilities Shatter Context Isolation
Electron patches 5 flaws (CVSS 7.0-8.4) allowing sandbox escapes and context isolation bypass. Update to v41, 40, or 39 to secure your desktop app now.
⤷ Title: Vercel Breach Linked to Context.ai, ShinyHunters Says It’s Not Involved
════════════════════════
𐀪 Author: Waqas
════════════════════════
ⴵ Time: Mon, 20 Apr 2026 21:21:48 +0000
════════════════════════
⌗ Tags: #Data Breaches #Security #BreachForums #Context.ai #Cyber Attack #Cyber Crime #Cybersecurity #data breach #ShinyHunters #Vercel
════════════════════════
𐀪 Author: Waqas
════════════════════════
ⴵ Time: Mon, 20 Apr 2026 21:21:48 +0000
════════════════════════
⌗ Tags: #Data Breaches #Security #BreachForums #Context.ai #Cyber Attack #Cyber Crime #Cybersecurity #data breach #ShinyHunters #Vercel
Hackread
Vercel Breach Linked to Context.ai, ShinyHunters Says It’s Not Involved
Vercel confirms a breach linked to Context.ai, while a hacker lists alleged data for $2M. ShinyHunters denies involvement in the incident.
⤷ Title: The AI Weak Link: How a Third-Party Breach Exposed Vercel Customer Secrets
════════════════════════
𐀪 Author: ddos
════════════════════════
ⴵ Time: Tue, 21 Apr 2026 09:17:43 +0000
════════════════════════
⌗ Tags: #Data Leak #API Keys #cloud security #Context.ai #Cybersecurity 2026 #data breach #Environment Variables #Mandiant #Next.js #OAuth Security #supply chain attack #Vercel
════════════════════════
𐀪 Author: ddos
════════════════════════
ⴵ Time: Tue, 21 Apr 2026 09:17:43 +0000
════════════════════════
⌗ Tags: #Data Leak #API Keys #cloud security #Context.ai #Cybersecurity 2026 #data breach #Environment Variables #Mandiant #Next.js #OAuth Security #supply chain attack #Vercel
Penetration Testing Tools
The AI Weak Link: How a Third-Party Breach Exposed Vercel Customer Secrets
The month of April concluded for the American firm Vercel with a distressing incident that precipitously transcended the
⤷ Title: Context Bombing: A Novel Defense Against AI Hackers
════════════════════════
𐀪 Author: Nam Phong
════════════════════════
ⴵ Time: Wed, 15 Jul 2026 06:53:12 +0000
════════════════════════
⌗ Tags: #Cybercriminals #AI security #AWS security #cloud defense #Context Bombing #Tracebit
════════════════════════
𐀪 Author: Nam Phong
════════════════════════
ⴵ Time: Wed, 15 Jul 2026 06:53:12 +0000
════════════════════════
⌗ Tags: #Cybercriminals #AI security #AWS security #cloud defense #Context Bombing #Tracebit
Information Security News
Context Bombing: A Novel Defense Against AI Hackers
Malicious commands hidden within text or files have served as tools to breach artificial intelligence systems for years. Now, defenders of these systems have begun utilizing the very same tactic t…
⤷ Title: Codex Cuts Its Context Window to 272K and Forbids rm -rf $HOME After Home-Directory Deletions
════════════════════════
𐀪 Author: Do Son
════════════════════════
ⴵ Time: Mon, 20 Jul 2026 05:23:45 +0000
════════════════════════
⌗ Tags: #Technology #AI safety #Codex #Context Window #GPT_5.6 #OpenAI
════════════════════════
𐀪 Author: Do Son
════════════════════════
ⴵ Time: Mon, 20 Jul 2026 05:23:45 +0000
════════════════════════
⌗ Tags: #Technology #AI safety #Codex #Context Window #GPT_5.6 #OpenAI
Daily CyberSecurity
Codex Cuts Its Context Window to 272K and Forbids rm -rf $HOME After Home-Directory Deletions
Because Codex is open source, we can watch the development team’s daily commits and spot notable changes in the code repository. Code committed two days ago reveals that the context window h…