AGENTS.md files help Codex and similar AI tools understand project-specific instructions, coding conventions, and organizational structures. They can reside anywhere within a Codex container, providing valuable context to AI agents. Like other project configuration files, these instructions are treated as trusted context by the agent. This trust model is by design, but it creates an interesting attack surface when a malicious dependency is able to write or modify these files at build time.
Mitigating Indirect AGENTS.md Injection Attacks in Agentic Environments | NVIDIA Technical BlogThis attack path highlights important considerations for the future of agent-assisted development. Extended supply chain risk: Traditional supply chain attacks focus on injecting malicious code directly. In agentic environments, a compromised dependency can also redirect the agent itself, extending familiar supply chain risks into a new dimension, such as injecting subtle delays that cause performance degradation or denial-of-service scenarios. Instruction following under adversarial conditions: When the agent followed injected configuration directives, including instructions to conceal its
Mitigating Indirect AGENTS.md Injection Attacks in Agentic Environments | NVIDIA Technical Blog… Discuss 0 Discuss 0 Tags Agentic AI / Generative AI | Trustworthy AI / Cybersecurity | General | Intermediate Technical | Deep dive | AI Agent | AI Red Team | featured | Security for AI | Trustworthy AI About the Authors About Daniel Teixeira Daniel Teixeira is a senior offensive security researche… …
… Long-running agents like OpenClaw have shown productivity gains but also pose security risks. Today’s agent runtimes resemble the early days of the web. They’re powerful but missing core security primitives: sandboxing, permissions, and isolation. …
… See the blog posts Sandboxing Agentic AI Workflows with WebAssembly and Practical Security Guidance for Sandboxing Agentic Workflows and Managing Execution Risk for more information about ways to secure agentic behavior. Security is a core concern, not an afterthought. …
… Discuss 1 Discuss 1 Tags Agentic AI / Generative AI | Trustworthy AI / Cybersecurity | General | Intermediate Technical | Deep dive | AI Agent | AI Red Team | featured About the Authors About Rich Harang Rich Harang is a Principal Security Architect at NVIDIA, specializing in ML/AI systems, with ov… …
… Discuss 0 Discuss 0 Tags Agentic AI / Generative AI | Developer Tools & Techniques | Trustworthy AI / Cybersecurity | General | Nemotron | Intermediate Technical | Tutorial | Build AI Agents | claws | DGX Spark | featured | LLMs | NemoClaw | OpenShell About the Authors About Patrick Moorhead Patric… …
… See NeMo Agent Toolkit in Action Create Your Own AI Agent 18:31 Benchmarking and Optimizing AI Agents 11:04 How To Develop Teams of AI Agents 23:06 Optimize Your AI Agent Workflows 05:29 How NeMo Agent Toolkit Works NVIDIA NeMo Agent Toolkit provides unified monitoring and optimization for AI agent… …
… Power physical AI agents with Jetson Modern physical AI agents are evolving rapidly with Gemma 4 models that integrate audio, multimodal perception, and deep reasoning capabilities. …
… Discuss 0 Discuss 0 Tags Agentic AI / Generative AI | Trustworthy AI / Cybersecurity | General | cuVS | NIM | Beginner Technical | Best practice | AI Red Team | DLI | featured | LLMs About the Authors About Rich Harang Rich Harang is a Principal Security Architect at NVIDIA, specializing in ML/AI s… …
… The open source Fleet Intelligence agent and the incorporation of cutting-edge integrity and attestation technologies underscore the NVIDIA commitment to transparency and security. …
… Security implications Constrained decoding changes the probability distribution of the model’s output before execution. That makes it useful, but only as one layer in an agentic AI control stack. The interesting security properties are: Reliability as a security property . …
