🛡️ Guardrails - alanxu.80

Miasma Worm Hits Microsoft Again: Azure Functions Action and 72 Other Repositories Disabled After Supply Chain Attack Targeting AI Coding Agents

🧠LLMs News

infoworld.com··Hacker News

ashp15205/guardian-runtime: A zero-latency, local-first runtime firewall for LLMs. Intercept every prompt and response locally to stop data leaks and runaway token costs.

🧩AI Frameworks Code

github.com··Hacker News, Hacker News

My side of the jqwik anti AI logging drama

🔐AI Security Blog

blog.johanneslink.net··Lobsters, Hacker News, Hacker News

Claude on Google Cloud: Monitoring and Securing Agents at Scale

🔐AI Security

anthropic.com·

Claude Fable 5 and new AI safety fables

🔐AI Security News

interconnects.ai··Hacker News

Risk Under Pressure: Compute-Aware Evaluation of Adversarial Robustness in Language Models

🧠LLMs Academic

arxiv.org·

The complete guide to llm and genai data security best practices

✍️Prompt Engineering Blog

dev.to··DEV

ShadowFeed Weekly #1: IronWorm npm Attack, $36M Humanity Protocol Hack, Microsoft Repos Compromised

🔐AI Security

x.com··DEV

Claude Fable is relentlessly proactive

🔐AI Security

simonwillison.net··Hacker News

Context-Fractured Decomposition Attacks on Tool-Using LLM Agents: Exploiting Artifact Provenance Gaps

🔭AI Observability Academic

arxiv.org·

Agent Series (13): Agent Security and Defense — Prompt Injection, Tool Abuse, and Data Leakage

🔐AI Security Blog

dev.to··DEV

Is security a skill issue? Five scanners, 3,084 skills, a different verdict 64% of the time

🔐AI Security

trymastro.com··Hacker News

inflightsec/agent-vault-proxy: Just-in-time API keys for AI agents - and any other process you route through it: the caller only ever sees a placeholder.

🔐AI Security Code

github.com··Hacker News

GitInject: Real-World Prompt Injection Attacks in AI-Powered CI/CD Pipelines

🔐AI Security Academic

arxiv.org·

If You Use Claude or Gemini, This Microsoft Breach Means Your Data Is at Risk

🔐AI Security

scienspire.com··Hacker News

Mini Shai-Hulud, Miasma, and Hades Worms Target Bioinformatics and MCP Developers via Malicious PyPI Wheels

🧩AI Frameworks Blog

socket.dev··Hacker News

Claude Fable 5 Was Jailbroken in 48 Hours. Here's What Actually Stopped Nothing.

✍️Prompt Engineering Blog

dev.to··DEV

PreAct-Bench: Benchmarking Predictive Monitoring in LLMs

🔭AI Observability Academic

arxiv.org·

Miasma Worm Hits Microsoft Again: Azure Functions Action and 72 Other Repositories Disabled After Supply Chain Attack Targeting AI Coding Agents

Google ADK Security: 5 Layers That Defend AI Agents From Prompt Injection

Meet Hades: The malware that lies to AI security agents

ashp15205/guardian-runtime: A zero-latency, local-first runtime firewall for LLMs. Intercept every prompt and response locally to stop data leaks and runaway token costs.

My side of the jqwik anti AI logging drama

Claude on Google Cloud: Monitoring and Securing Agents at Scale

Claude Fable 5 and new AI safety fables

Risk Under Pressure: Compute-Aware Evaluation of Adversarial Robustness in Language Models

The complete guide to llm and genai data security best practices

ShadowFeed Weekly #1: IronWorm npm Attack, $36M Humanity Protocol Hack, Microsoft Repos Compromised

Claude Fable is relentlessly proactive

Context-Fractured Decomposition Attacks on Tool-Using LLM Agents: Exploiting Artifact Provenance Gaps

Agent Series (13): Agent Security and Defense — Prompt Injection, Tool Abuse, and Data Leakage

Is security a skill issue? Five scanners, 3,084 skills, a different verdict 64% of the time

inflightsec/agent-vault-proxy: Just-in-time API keys for AI agents - and any other process you route through it: the caller only ever sees a placeholder.

GitInject: Real-World Prompt Injection Attacks in AI-Powered CI/CD Pipelines

If You Use Claude or Gemini, This Microsoft Breach Means Your Data Is at Risk

Mini Shai-Hulud, Miasma, and Hades Worms Target Bioinformatics and MCP Developers via Malicious PyPI Wheels

Claude Fable 5 Was Jailbroken in 48 Hours. Here's What Actually Stopped Nothing.

PreAct-Bench: Benchmarking Predictive Monitoring in LLMs