Black Hat · August 29, 2026

Black Hat Asia 2026 | Bad Vibes - Pwning Coding Agents 70 Times With The Same Bugs

video Agent Security AI Red Teaming Prompt Injection AI Engineering

Black Hat Asia 2026 | Bad Vibes - Pwning Coding Agents 70 Times With The Same Bugs video thumbnail

Why it matters

Palo Alto Networks researchers found the same command-parser, protected-path, and sandbox-boundary failures across major coding agents. Their survey produced more than 81 vendor reports and 18 assigned or reserved CVEs, including allowlist bypasses through compound shell syntax, path-equivalence errors, unsafe moves and symlinks, and gaps between file and terminal controls.

My takeaway: Run coding agents in an isolated container or VM and enforce one policy across terminal and file tools. Parse the complete command using the target shell, canonicalize resolved paths and nonexistent descendants, mediate moves and symlinks, and regression-test compound commands, alternate path forms, configuration writes, symlink races, and prompt-injection-driven execution.

Keep exploring

More curated notes connected through Agent Security and AI Red Teaming.

OpenAI News · framework

Pacing model development in an era of cyber-critical capabilities

OpenAI says preliminary evidence that Astra may meet its Critical cybersecurity threshold led it to pause frontier reinforcement-learning work for two weeks and keep its largest planned run on hold. New safeguards include stronger workload and network isolation, continuous boundary testing, token-level monitoring that escalates suspicious tool activity, and broader alignment checks for deception, reward hacking, and unauthorized access.

OWASP GenAI Security Project · guide

OWASP Top 10 for Agentic Applications for 2026

OWASP's community guide organizes agentic-system risk into ten categories, including goal hijacking, tool misuse, identity and privilege abuse, memory poisoning, insecure inter-agent communication, cascading failures, and rogue-agent behavior. It provides a shared taxonomy and mitigation starting point rather than a certification checklist or evidence that a deployed system is secure.

OpenAI News · framework

A blueprint for democratic governance of frontier AI

OpenAI proposes a three-part U.S. frontier-AI governance model: harmonize emerging state safety laws into a federal baseline, strengthen CAISI as an evaluation and standards institution, and coordinate a broader resilience program. Proposed controls include severe-risk evaluations, transparency reports, independent audits, safety-incident reporting, model-weight security, whistleblower protection, and periodic technical assessments.