300 Agents, 60 People: Runlayer Found Engineer Details Real-World Agent Security Failures
AI Engineer · youtube · 2026-10-27
Runlayer founding engineer Aiden Sochowski shares what a gateway with full session and tool-call logging revealed at his 60-person company running 300+ agents daily: prompt injection via email, PR-review agents talked into approvals, over-permissioned bots, one poisoned agent contaminating others, buggy vibe-coded MCP servers burning tokens, and a model switch that blew up costs over 100x in a week. His core argument: most companies have zero visibility into their agents, and session-level logs are the foundation for both security auditing and scoping agents down.
More from coding & agent
- Testing Grok Bot: agent demands full-device GitHub PAT despite existing repo access — vincent_koc · 2026-10-27
- OpenCode v2.2.0 adds desktop plugins and cache-friendly keyed session instructions — github-actions[bot] · 2026-10-27
- Fresh chat per task: 7 context engineering tips to avoid context rot — every · 2026-10-27
- Redditor open-sources a simple workflow for infinite lipsynced talking head videos with MiniMax H3 — martinerous · 2026-10-27
- System prompts aren't a security boundary: Reddit debates how much access AI agents should get — Putrid_Ad734 · 2026-10-27
- AI told to install Minecraft mods starts building its own C++ game engine — Promptmethus · 2026-10-27