Find notable cyber news and cases, enriched with sources, timelines, and signals.
Home Watchlists Beta Browse News Feed Stats About Contact

AI browsers indirect prompt injection via URL fragments HashJack security flaw

Vulnerability
First reported
Last updated
Happening score
H score 28
2 unique sources, 2 articles

Summary

Hide ▲

HashJack is an indirect prompt injection vulnerability in AI browsers that hides attacker instructions after the # symbol in legitimate URLs, letting a normal-looking link manipulate the assistant into unsafe actions. The issue affected Comet, Copilot for Edge, and Gemini for Chrome; by November 25, fixes had been applied to Comet and Copilot for Edge, while Gemini for Chrome remained unresolved. Researchers said the technique could enable callback phishing, data exfiltration, misinformation injection, malicious downloads, opening ports, and credential theft, and that ordinary server-side and network defenses do not see the fragment payload.

Related Happenings

Claude Chrome forged-click and skipPermissions security flaw

Vulnerability
H score35 First: 14.07.2026 20:27 Last: 14.07.2026 20:27 Sources 1

About this happening: Claude for Chrome still accepts synthetic clicks on its onboarding button, letting a rogue extension trigger allowlisted tasks for Gmail, Google Docs, and Calendar...

Open Happening

Search for perplexity ai malicious Chrome extension

Malware Activity
H score29 First: 29.06.2026 21:40 Last: 29.06.2026 21:40 Sources 1

About this happening: A malicious Chrome extension named Search for perplexity ai impersonated Perplexity AI while intercepting search traffic and collecting browsing information th...

Open Happening

LayerX BioShocking prompt injection against agentic browsers

Technical Analysis
H score30 First: 24.06.2026 19:05 Last: 24.06.2026 19:05 Sources 1

About this happening: Researchers demonstrated BioShocking, a prompt-injection technique that pushed six agentic browsers and plugins past guardrails and made them copy login credentials fo...

Open Happening

OpenAI ChatGPT renderer Markdown link/image phishing security flaw

Vulnerability
H score16 First: 29.05.2026 21:07 Last: 29.05.2026 21:07 Sources 1

About this happening: ChatGPT has a response-renderer vulnerability that turns summarized third-party pages into live phishing links and auto-fetched attacker-hosted images inside the t...

Open Happening

Torg Grabber browser-extension theft activity

Malware Activity
H score36 First: 25.03.2026 20:32 Last: 25.03.2026 20:32 Sources 1

About this happening: The Torg Grabber infostealer is actively stealing data from 850 browser extensions, including 728 cryptocurrency wallet extensions, which raises the risk of account ta...

Open Happening

Timeline

  1. 26.11.2025 12:15 2 articles · 7mo ago

    HashJack fixes applied to Comet and Copilot for Edge

    Mitigation Patch Update

    Perplexity and Microsoft had applied fixes for Comet and Copilot for Edge by November 25, while Gemini for Chrome remained unresolved for HashJack, the indirect prompt injection that hides malicious instructions in the text after the # symbol in legitimate URLs.

    Show sources
    Open in new tab
  2. 26.11.2025 12:15 2 articles · 7mo ago

    Security researchers disclose HashJack indirect prompt injection

    Initial Disclosure

    Security researchers disclosed HashJack as a new indirect prompt injection vulnerability that tricks AI browsers such as Comet, Copilot for Edge and Gemini for Chrome by hiding malicious instructions in URL fragments after the # symbol. The technique can support callback phishing, data exfiltration, misinformation injection, malicious downloads, opening ports and credential theft, while traditional network and server defenses like intrusion detection systems do not see the fragment payload.

    Show sources
    Open in new tab