AgentsContextCrush: MCP Documentation Poisoning via Context7 Custom RulesNoma Security·high signalXBlueskyLinkedInCopy linkAnyone could inject malicious rules into Context7 MCP server. Zero sanitization. npm typosquatting equivalent for MCP ecosystem.SourceSource pageNoma Security↳ Follow the threadPolicy dependency / Stack layerCROSS-CATEGORY: Five Vendors in Ten Days Shipped the Same Product — a Gateway That Sits Between Your Apps and Your Agents and Meters ThemHelp Net Security (A10, SelectHub; corroborated by Snowflake's engineering blog, GlobeNewswire/AI-R Aug 10, and harnessrouter.ai)Stack layer / Threat patternllmfit v1.1.10 Adds RamaLama Runtime Discovery and Publishes the First MLX vs llama.cpp Metal Head-to-Head on Identical HardwareGitHubStack layer / Threat patternColoring Positive Words Green Shifts VLM Sentiment Predictions — a Styling-Only Prompt Injection ChannelarXiv 2608.14286Policy dependency / Stack layerShow HN: PyScrappy Pairs Self-Healing Scraper Selectors With an MCP Server So Agents Can Re-Target Broken PagesGitHub / Hacker NewsStack layer / Threat patternVendo Launched an Embedded Layer That Lets Your Own Users Build Features Inside Your SaaS Product Without CodeProduct Hunt (Aug 17 daily leaderboard; single sourceStack layer / Threat patternoMLX 0.6.0 Ships Distributed LLM Serving Across Multiple Macs — Qwen3.6-27B Goes 16.1 → 28.6 tok/s on Two MachinesGitHubPolicy dependency / Stack layercode-graph-rag v0.0.639 Adds Runtime Call Tracing for Seven Language Runtimes and Interprocedural Taint PropagationGitHubPolicy dependency / Stack layerIn the Same Post, Amodei Rejects the Regulation-vs-Open-Models Binary and Endorses a FINRA-Style Self-Regulator for AITechCrunch