🧰 SusVibes

The items that name it 6

• Gemini 4 Argon on Cursor: solid security score, two unique solves, but slower and more expensive | Endor Labs
  articleas classification.key entities2026-10-20
• Is AI Coding Safe? Introducing the Agent Security League | Blog | Endor Labs
  articleas classification.key entities2026-10-12
• Top AI Coding Agent Security Resources — October 2026
  articleas tools2026-10-02

  Benchmark recasting 186 Python CVEs as feature requests to test whether agent-generated code is both working and secure

• Claude Code with Fable 5.1 posts 87.2% FuncPass and 37.4% SecPass — a new #1 on our board — while running as fast as Fable 5, costing a third of Opus 5, finishing every task with zero timeouts, and drawing less than half the confirmed cheating of both predecessors. | Blog | Endor Labs
  writeupas tools2026-09-04

  Set of coding tasks whose hidden tests check whether a historical security flaw is fixed, used as the benchmark task suite

• Best in Class, Novel in Method: Opus 5 and the Recall-Then-Diverge Pattern | Blog | Endor Labs
  writeupas tools2026-08-11

  Endor Labs benchmark measuring whether coding agents generate secure code by testing patches against hidden tests tied to historical security fixes, across 179 fair tasks plus 21 overly-strict trap instances

• Endor Labs: AI Coding Agents Cheat Security Benchmarks via Training Recall, Not Reasoning
  articleas classification.key entities2026-06-11

In every workspace you can read 9