https://hackyjs.com/posts/anatomy-of-the-meta-muse-0-day-how-prompt-injection-becomes-a-hijacked-tool-call Hacky JS en 2026-09-22T00:00:00.000Z Anatomy of the Meta Muse 0-Day: How Prompt Injection Becomes a Hijacked Tool Call ai-agents, security, prompt-injection, tool-calling, meta-muse https://hackyjs.com/posts/auditing-ai-agent-tool-execution-on-macos-after-the-meta-muse-zero-day Hacky JS en 2026-09-22T00:00:00.000Z Auditing AI Agent Tool Execution on macOS After the Meta Muse Zero-Day ai-agent-security, macos, zero-day, sandboxing, malware-injection https://hackyjs.com/posts/auditing-heif-decoding-in-server-side-javascript-after-the-heif-heist Hacky JS en 2026-09-22T00:00:00.000Z Auditing HEIF Decoding in Server-Side JavaScript After the HEIF Heist heif, image-parsing, rce, nodejs, security https://hackyjs.com/posts/building-a-cost-per-task-ledger-for-agent-pipelines-on-gpt-6-opus-55-and-grok-47 Hacky JS en 2026-09-22T00:00:00.000Z Building a Cost-Per-Task Ledger for Agent Pipelines on GPT-6, Opus 5.5, and Grok 4.7 llm-costs, agent-pipelines, inference-economics, prompt-caching, api-pricing https://hackyjs.com/posts/lifecycle-script-diffing-and-provenance-checks-hardening-npm-releases-after-the-65-repo-breach Hacky JS en 2026-09-22T00:00:00.000Z Lifecycle-Script Diffing and Provenance Checks: Hardening npm Releases After the 65-Repo Breach npm, supply-chain-security, ci-cd, provenance, github-actions https://hackyjs.com/posts/porting-a-real-python-backend-to-cloudflare-workers-what-pyodidewasm-actually-breaks Hacky JS en 2026-09-22T00:00:00.000Z Porting a Real Python Backend to Cloudflare Workers: What Pyodide/WASM Actually Breaks cloudflare-workers, python, pyodide, wasm, edge-computing