Background

Anthropic Addresses High-Risk Vulnerabilities in Git Server

Article arrow_drop_down
Anthropic Addresses High-Risk Vulnerabilities in Git Server
Key Points:
  • An attacker could exploit Git server vulnerabilities.
  • No crypto impact or asset movement reported.
  • Security experts recommend immediate patch updates.

Cyata Security researchers discovered and disclosed three high-risk vulnerabilities in Anthropic’s MCP Git server, leading to patches for arbitrary file access and remote code execution. These fixes occurred quietly without public acknowledgment from Anthropic’s leadership.

Anthropic has rectified three high-risk vulnerabilities in its MCP Git server, initially disclosed by Cyata Security researchers in 2025. Despite the fix, no responses have been received from Anthropic’s leadership on this matter.

Experts highlight the significance of Anthropic’s patch, correcting vulnerabilities without affecting cryptocurrency markets or financial sectors.

Anthropic corrected multiple security vulnerabilities in its MCP Git server. The issues were discovered by Cyata Security and patched by December 2025. These vulnerabilities included arbitrary file access and remote code execution risks.

The vulnerabilities, tracked as CVE-2025-68143, CVE-2025-68144, and CVE-2025-68145, were identified by Cyata. No cryptocurrency was affected despite potential risks, and no direct public acknowledgment from Anthropic followed the patching.

Experts conveyed no financial repercussions on crypto markets.

However, the vulnerabilities highlight critical security considerations for digital infrastructures. Shahar Tal, Cyata CEO, emphasized the need for enhanced security measures across the board.

Potential outcomes indicate the vulnerabilities serve as a wake-up call on digital security frameworks. Yarden Porat, researcher at Cyata, stressed that organizations must anticipate more complex system interactions to mitigate risks effectively, reinforcing the industry’s ongoing examination of security vulnerabilities.

“These flaws can be exploited through prompt injection, meaning an attacker who can influence what an AI assistant reads (a malicious README, a poisoned issue description, a compromised webpage) can weaponize these vulnerabilities without any direct access to the victim’s system.” — Yarden Porat, Researcher, Cyata Security

Wake-Up Call for Digital Security: Anthropic’s Git Server Issues

About the author

About the author call_made

CoinLineup Editorial Team

The CoinLineup Editorial Team comprises experienced financial analysts and cryptocurrency researchers dedicated to delivering accurate, timely market intelligence. Our editors verify all data against primary sources including SEC filings, central bank reports, and on-chain analytics before publication.

More posts Follow

Related

About Coinlineup

CoinLineup is a specialized platform dedicated to empowering investors with the knowledge and tools needed to succeed in both the financial stock market and the crypto market. Our primary focus is to provide comprehensive market insights by delivering real-time and historical data, solid investment strategies, and trading tips. We aim to equip investors with accurate information, allowing them to make well-informed decisions in their financial endeavors.

Copyright 2024 coinlineup.com. Crypto, Stocks, and Forex – All in One Place.

Login to enjoy full advantages

Please login or subscribe to continue.

âś–

Go Premium!

Enjoy the full advantage of the premium access.

Login

âś–

Stop following

Unfollow Cancel

âś–

Cancel subscription

Are you sure you want to cancel your subscription? You will lose your Premium access and stored playlists.

Go back Confirm cancellation

âś–