# Security: what I've logged

> Security comes up in 2 posts of my AI log, from October 2026 to October 2026, and each item below links to the primary source. I filed these posts under frontier, agents and coding and open weights.

## Posts

- 2026-10-22 [OpenAI CLI, Omnara, PHP2Shell, KLS proof, RFD4-Proteina](https://heyaliasif.com/ai/2026-10-22-openai-cli-omnara-php2shell-kls-rfd4-proteina)
- 2026-10-21 [Claude Duet, Granite Speech 5, Kumo Relational, attack surface](https://heyaliasif.com/ai/2026-10-21-claude-duet-granite-speech-5-kumo-attack-surface)

Canonical: https://heyaliasif.com/ai/topics/security