AINewsnow

howbad: What a Prompt-Injected Coding Agent Could Do With the Credentials on Your Machine

Claude Code, Codex and Cursor run as me. A prompt injection in a README or an issue comment doesn't need a sandbox escape. It gets whatever my shell gets: the gh token, the npm token that publishes packages, the AWS profile from this morning, the Slack token pasted into an MCP config. I knew roughl…

Read the full story at DEV Community — AI ↗

Timeline · 1 report

  1. 2026-11-14 21:50 · DEV Community — AI
    howbad: What a Prompt-Injected Coding Agent Could Do With the Credentials on Your Machine

More stories

  1. Nvidia board to decide Jensen Huang’s fate — Financial Times AI
  2. Shipping a remote MCP server with OAuth: what the clients actually check — DEV Community — AI
  3. What AI coding tools actually cost in November 2026 (16 tools, list prices with sources) — DEV Community — AI
  4. Offering AI coding governance to your clients: an MSP and vCISO playbook — DEV Community — AI
  5. trust-me-bro: Reading an Agent Skill Before It Reads Your .env — DEV Community — AI
  6. Cursor confirms breach enabled malicious Claude Code extension in marketplace — BleepingComputer
  7. The git worktrees your coding agents leave behind, and how to clean them up — DEV Community — AI
  8. Top 25 Open-Source Model Context Protocol (MCP) Servers for Claude Desktop & Cursor (2026) — DEV Community — AI

Get the daily brief of stories like this at 6:30 every morning →