A malicious repo downgraded Claude Code’s review from Fable to Haiku, steered the agent away from the payload, and turned a routine test run into malware execution.
Fable to Haiku: How a Malicious Repo Tricked Claude Code Into Running Malware | Straiker
About this summary. This is a short, independently written summary of an article first published by Straiker. Cyber Security News did not report or verify the underlying story. Read the original: https://www.straiker.ai/blog/how-a-malicious-repo-tricked-claude-code-into-running-malware
Source attribution: headline and facts are from Straiker (straiker.ai). Summary method: excerpt of the source description. See our source attribution policy.
Related coverage
Canadian regulator opens probe of IDScan for allegedly violating data privacy laws
The investigation, announced Monday, will probe IDScan’s security practices and whether victim notifications were adequate under Canada’s federal private-sector privacy…
AL26-022 - Vulnerability impacting F5 BIG-IP Access Policy Manager (APM) – CVE-2026-94127
Number: AL26-022 Date: September 22, 2026 Audience This Alert is intended for IT professionals and managers. Purpose An Alert is used to raise awareness of a recently…
🏴☠️ Shinyhunters has just published a new victim : PSA - READ THIS NOW
Dear Assistant Director Brett Leatherman of the FBI Cyber Division & Director Kash Patel of the FBI, During Quarter Two of this year the Federal Bureau of Investigation…
Hacking group ShinyHunters claims it breached the FBI, stole agents’ and applicants’ data
The theft of agents' personal information could present a major counterintelligence threat, where agents and their families are extorted into cooperating with a foreign…
CVE-2026-95831: Crypt::SelfCertificate versions from 1.01 through 1.05 for Perl contains malware which executes Python code from an obfuscated URL
Posted by Robert Rothenberg on Sep 22 ======================================================================== CVE-2026-95831 CPAN Security Group…
New ClosedQuorum Windows malware uses AI for attack decisions
A new Windows malware named ClosedQuorum uses Google Gemini, DeepSeek, Qwen, and Mistral AI models to autonomously determine the actions to take during post-compromise…