SShortSingh.
TopTechnologyAI & MLStartupsProgrammingScienceBusinessFinanceCrypto & Web3World
Back to feed
Programming

AI coding agents can inadvertently leak secrets; new open-source tool aims to block it.

0
·1 views

AI-powered coding assistants like Claude Code and Cursor routinely read a project's .env file for context. A security risk arises when these agents, following seemingly innocent instructions, can write those secrets to other files and commit them to public repositories. A developer has created StingLabs, a free and open-source tool designed to prevent such leaks. It tracks data from its origin and blocks marked secrets from being transmitted via commands like git push or curl. The tool runs locally using deterministic rules and is available for macOS and Windows.

Read the full story at DEV Community

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

Log in to join the discussion and vote.

Log in

Related stories

0
ProgrammingDEV Community ·

Developer creates offline AI break planner for short outdoor activities

A developer has built an AI-powered application called Step Out that generates personalized outdoor break plans. The tool uses Google's Gemma 3 1B model running locally via Ollama on a user's computer, requiring no cloud connection or API fees. Users input their available time, location type, goal, and mobility preference to receive a customized activity schedule. The application operates entirely offline after initial setup, with no account creation or location tracking required. It was tested successfully on an Apple M1 laptop, generating activities like sound observation breaks in residential settings.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Technical guide resolves authentication conflicts between Confluent Schema Registry and MSK Serverless on EKS

A developer documented a multi-layer authentication conflict when deploying Confluent Schema Registry on EKS against an MSK Serverless cluster. The first issue involved Kubernetes injecting malformed environment variables, crashing the Schema Registry pod, which was fixed by disabling service link injection. A second problem was the Confluent base image lacking the required AWS MSK IAM authentication library for MSK Serverless. The solution required building a custom Docker image that manually adds the necessary JAR file to a specific directory.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Hacktoberfest Week 2 Challenge: Build Open-Source AI Side Hustles for Prizes

The second week of Hacktoberfest's five-week DEV Challenges began on October 11, 2026, running until October 18. Participants are tasked with building new projects using open-source AI or open-weight models that solve real-world problems for potential users. The challenge will select one overall winner and 17 category winners, with a total prize pool of $2,550 distributed among 18 winners. Projects should demonstrate how the open-source approach provides advantages over closed systems, such as cost savings or data privacy.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Over 21,000 unpatched Exchange servers expose critical SYSTEM-level flaw

A critical security vulnerability in Microsoft Exchange servers allows attackers to bypass authentication and achieve SYSTEM-level remote code execution. The flaw, tracked as CVE-2026-62911, affects Exchange Server 2016, 2019, and SE versions. In September 2026, over 21,800 internet-facing servers remained unpatched, primarily in the United States and Germany. While fixes exist for supported versions, many servers, especially outdated ones, lack a straightforward update path. Security experts recommend immediate patching or restricting internet access to the vulnerable MRSProxy component.

0 comments Read more at DEV Community
AI coding agents can inadvertently leak secrets; new open-source tool aims to block it. · ShortSingh