Truffle Launches Auto-Revocation for 1.4M Exposed Secrets
Original source
Nearly 1.5 million exposed secrets targeted in new revocation program
BleepingComputer →Truffle Security identified 1.4 million valid API credentials publicly exposed across repositories, packages, and container images. Since 2022, the company cataloged 543,000 credentials in GitHub repositories and over 431,000 AWS secrets scattered across development infrastructure, container registries, and CI logs. Most exposures stem from developer carelessness—credentials accidentally committed to public repos, included in published Docker images, or left in configuration files.
Developers routinely ignore breach notifications: 81% of reported keys remained functional six months after disclosure. This passive stance leaves downstream users and customers vulnerable. The problem has worsened since early 2025, when AI coding tools began embedding 2,655 working keys in their configuration files—an issue absent the previous year. Attackers have capitalized on this trend; in July, OpenAI documented AI agents exploiting exposed Hugging Face tokens to steal platform data.
Truffle’s new Secrets Revocation Program bypasses unresponsive developers by partnering directly with API issuers to automatically invalidate leaked keys within 14 days. Sixteen organizations—including Hugging Face, Notion, Vercel, and Twilio—have already enrolled. The approach builds on Google’s proven results: when Google switched from notifications to automatic disabling of compromised Cloud keys in June 2024, active exposed credentials dropped from 72% to 11%.
Read the full article
Continue reading at BleepingComputer →This is an AI-generated summary. Read the original for the full story.