Swarms adds 0-100 Security Scores to every marketplace prompt via SkillScanner
KyeGomezB · x · 2026-10-02
Swarms now shows a Security Score (0-100) and a Security tab on every prompt page in its marketplace, powered by its open-source scanner SkillScanner.
- Why: a prompt is a program in plain language — pasted into an agent with shell, browser, or API keys, one easy-to-miss sentence can make it read /.ssh/idrsa and exfiltrate it.
- Until now, the only way to vet a prompt was reading every line yourself.
- Each page now shows a score badge (green shield = SAFE), a recommendation, a verdict, and every underlying finding.
- The post explains the scoring, how scans work, and how to use it before running or publishing a prompt.
More from coding & agent
- Meta's new SWE-sweep eval makes coding agents find bugs in real repos with no hints — jack_w_rae · 2026-10-02
- Kaigen Engine opens closed beta: C-based cross-platform engine built for AI coding — gdechichi · 2026-10-02
- Kaigen open-sources Boids demo porting Unity's ECS sample to its C-based engine — gdechichi · 2026-10-02
- AI-native observability needs new SLIs beyond latency and error rate — rseroter · 2026-10-02
- Team of 8 AI agents beats best solo agent building Colosseum in Minecraft (0.72 vs 0.58) — DimitrisPapail · 2026-10-02
- AC2 adds automated reward-hacking monitors, AI agent investigates cheating RL runs — ypatil125 · 2026-10-02