Storia in 2 fonti

Coding Agent Horror Stories: The 29 Million Secret Problem | Docker

Learn how AI coding agents can expose credentials in supply chain attacks and how Docker Sandboxes keep secrets out of an agent's reach.

Raccontata da

docker.com

dev.to

Confronto fonti

2 prospettive sulla stessa storia

AI · summaries

docker.comStai leggendo14 h fa

Coding Agent Horror Stories: The 29 Million Secret Problem | Docker

Malicious Nx packages exploited AI agents via --dangerously-skip-permissions, driving 28.65M secrets leaked in 2025. AI-assisted code leaks credentials at 2x baseline—teams must sandbox agents and restrict permission-bypass flags.

originale

dev.to3 h fa

Docker returns to its coding-agent series with an argument shaped like a CI problem: no layer between the…

Docker warns that AI coding agents run with developer filesystem and credential access, with no sandbox isolation. For infrastructure teams, this mirrors self-hosted CI runners—sandbox solutions include containers, microVMs, or task-scoped isolation.

Leggi questa versione →

Coding Agent Horror Stories: The 29 Million Secret Problem | Docker

Confronto fonti

Coding Agent Horror Stories: The 29 Million Secret Problem | Docker

Docker returns to its coding-agent series with an argument shaped like a CI problem: no layer between the…

Timeline cronologica

Coding Agent Horror Stories: The 29 Million Secret Problem | Docker

Docker returns to its coding-agent series with an argument shaped like a CI problem: no layer between the agent and the host