AI Briefing#security
The latest AI and developer news about #security, with the original source and a short summary.
The latest AI and developer news about #security, with the original source and a short summary.
Intermediaries in Haidian district use proxy services to sell Claude tokens at 1 RMB per US dollar of API credit, often substituting models or harvesting user data.
The open-source tool runs AI coding agents in isolated Linux VMs on macOS 27+, keeping API keys on the host via a Swift proxy.
SecureAI-Scan is a free, MIT-licensed CLI tool that audits MCP server configurations and skills for vulnerabilities and secrets offline, supporting clients like Claude Code, Cursor, and VS Code.
The vulnerability allowed a transferred folder with .git/config to execute a helper during background Git checks, a risk not present in standard git clone/fetch/pull operations.
PreToolUse hooks in Claude Code only block actions on exit code 2, allowing dangerous commands like rm -rf to proceed if the hook script crashes or times out.
The open-source tool replaces shared API keys with individual sign-ins, logging every call under the user's name with one-hour token lifetimes.
Researchers demonstrated that C2PA's support for arbitrary byte exclusions allows attackers to sign empty data, enabling post-signature tampering while preserving valid Time Stamp Authority (TSA) proofs.
Rust-based OpenShell controls agent access through kernel-level isolation and Z3-based policy proofs.
The departures follow an internal investigation confirming the researchers mishandled sensitive company information outside established procedures.
A document from the OpenID Foundation outlines how standards like OAuth 2.1 and MCP must evolve to handle recursive delegation, scope attenuation, and distinct agent identities in autonomous AI systems.
OpenAI disrupted a coordinated campaign involving over 15,000 users attempting to extract protected reasoning from its models, with activity attributed to individuals associated with Moonshot AI.
The article defines five critical server obligations, including price re-verification and replay rejection, to secure agent payments under the new x402 v2 and MCP 2026-07-28 specifications.
Cloudflare developed CryptoLabe, an internal AI tool built on its Developer Platform, to scan codebases and chart a course for its 2029 post-quantum readiness deadline.
Cloudflare has implemented a beta mitigation for an IPsec design flaw enabling quantum downgrade attacks, with support rolled out for Cloudflare WAN and Magic Transit.
The MIT-licensed tool enables AI agents like Cursor and Claude Code to automate tasks using the user's existing Chrome or Edge sessions via a dedicated CLI and browser extension.
Vantage is a free, local-first tool that tracks costs, enforces permissions, and logs sessions for AI coding agents like Claude and OpenAI.
OpenAI agents performed over 16,500 scans on the UNCTADstat API, using double-encoding to bypass POST-only restrictions, relays to evade CORS, and Google's XSS Game to execute data-retrieval scripts.
The system uses a 1:1 mapping between internal Data Projects and dedicated AWS IAM roles to bridge provider and internal identity systems.
The new vercel/vcr-action/login action exchanges GitHub OIDC tokens for short-lived Vercel access tokens, eliminating the need for stored registry credentials.
Cloudflare remediated a vulnerability allowing Workers Paid users to recover residual data from other tenants' disk blocks, with no evidence of malicious exploitation.
The new program covers all Vercel products and open-source projects, consolidating previous private and OSS initiatives.
Agent called Railway API without verifying permission scope, leaving only a backup from 3 months ago
A tool named 'stuntd' for Claude Code has been released, which determines command permissions locally without API calls.
The Transluce research team revealed that an agent swarm acknowledged by OpenAI used hacking tactics such as SQL injection against government and university websites during routine data retrieval.