Researchers at Adversa AI disclosed a new attack technique called Cryptographic Context Injection that targets Grok, with a similar variant shown against Gemini. The attack embeds an encrypted payload…
Sentinel, an AI-powered penetration testing tool, claims to successfully hack into systems and gain root access in 9 out of 10 attempts, according to its product description. The tool uses a persisten…
Researchers disclosed in August 2026 that encrypted reasoning objects across OpenAI, Anthropic, and Google APIs could be replayed outside their original context, allowing secrets to be extracted acros…
An engineer built Sentinel, an autonomous real-time Voice AI emergency dispatcher, in 10 days during the #VoiceForBharat challenge. The system uses Deepgram Nova-3 for speech-to-text, Google Gemini fo…
OpenAI paused work on its Astra model after evaluations suggested it may have reached 'Critical' cyber capability, including potential autonomous zero-day exploitation. Meanwhile, agentic models from …
Mnemara, a Claude conversation runtime that re-reads a role document on every turn and pins it as the system prompt, reported zero out-of-lane writes and zero protocol violations in a 101-turn coding …
Researchers disclosed the 'PleaseFix' issue, showing that AI browser agents can be hijacked by malicious instructions hidden in web content, leading to zero-click attacks. The vulnerability is structu…
AliceLabs LLC launched MarketNow, an npm registry hosting 9,248 MCP servers for AI agents like Claude Desktop, Cursor, Cline, Continue, and Aider, with 43 free skills and paid options from $0.99 to $9…
The town of Jay, Maine, population 4,600, unanimously supported a proposed AI data center on the site of the former Androscoggin Mill, which exploded in 2020, but the developer Tony McDonald's project…
A researcher disclosed that Microsoft 365 Copilot for Word can be manipulated by hidden, invisible-formatted text embedded in a document, causing it to alter financial figures and propagate the malici…
Sentinel, a new open-source tool, detects breaking changes in vendor API changelogs before they affect customer codebases, scans repositories for affected usages, and opens a pull request with fix dif…
A developer known as Skull Skills has released a framework called Skull that orchestrates multi-agent AI teams within a local environment using markdown files, without requiring API keys or external a…
A developer has released Skull Skills, a free, open-source tool that turns Claude Code into a multi-agent team. By saying "skull," a leader interviews the user, maps the project, and assembles special…
Dark Reading covered research showing that LLM safety guardrails are not applied consistently across languages, creating a multilingual jailbreak gap. A prompt blocked in English can bypass the same m…
Edison Flores of AliceLabs LLC built Agent Trust Card (ATC), an SSL-like trust layer for AI agents, to solve the verification problem when agents call each other autonomously. The system uses an 8-lay…
Sentinel, an open-source QA agent released under MIT, reads a full-stack codebase to derive and test critical business flows end to end, surfacing bugs invisible from the UI alone. When pointed at Kar…
A developer renamed their protocol from ACP to ATC (Agent Trust Card) after discovering IBM/BeeAI's existing ACP protocol. ATC is described as an identity, trust, encryption, and payment layer for AI …
A developer built ATC (Agent Trust Card), a system that provides AI agents with cryptographic identity, security audits, encryption, protocol translation, and autonomous payments. ATC aims to solve th…
Sentinel's SENT token surged 20% to $0.017 after Robinhood listed the asset for spot trading, with 24-hour volume reaching $143 million. The AI-focused token's market cap rose to $130 million as retai…
Researchers reported in July 2026 that nine widely used AI coding tools, including Cursor, GitHub Copilot, Gemini CLI, and Windsurf, share a common failure mode called HalluSquatting. Attackers exploi…