1. Paolo Costanzo Sep 29, 2026
   

   Vulnerability

   In six of the ten AI agent flaws examined, the model's judgement was never the decisive factor.

   • The ten disclosures, published between 26 June and 24 September 2026, are classified into four classes according to the trigger point.

   • In the other four cases, the model was manipulated, but permissions and configuration decided the extent of the damage.

   • GitSpawn groups eight findings in seven command-line agents, including Claude Code, Goose, Codex and Cursor.

   • A correction on 13 October brings the figure from seven to six, after the reclassification of CVE-2026-55607 (Claude Code).

   Read the original article Outside the Prompt: What This Summer's AI Agent Vulnerabilities Have in Common paolocostanzo.github.io

Back to the week