Twenty GitHub advisories landed on PraisonAI in one day, and most of the auth silently does nothing.

On August 25 the advisory database published 20 advisories against PraisonAI and praisonaiagents: 1 critical, 15 high, 4 medium. The recurring pattern is authentication that's declared but never enforced. praisonai serve agents --api-key is silently ignored. AgentServer declares auth_token and enforces it on no route. The Recipe server fails open when API-key or JWT auth is configured. The async Jobs API /api/v1/runs has no authentication at all (GitHub Advisory). The rest cover SSRF via DNS rebinding and redirect-following, workspace escape via symlinks, and an ast_grep_rewrite tool that rewrites arbitrary files without its @require_approval gate.