3 commits

Author	SHA1	Message	Date
rcourtman	3e3de5325b	security: allow rm on /var/tmp and /tmp with approval ... Updated command policy to be more nuanced: BLOCKED (hard block, never allowed): - rm -rf / (root) - rm -rf /* (root wildcard) - rm -rf /home, /etc, /usr, /var/lib, /boot, /root, /bin, /sbin, /lib, /opt REQUIRE APPROVAL (user must click 'Run'): - rm -rf /var/tmp/* (Proxmox vzdump temp files) - rm -rf /tmp/* This allows AI to suggest cleaning up vzdump temp files while still protecting against destructive operations on critical paths.	2025-12-21 18:53:08 +00:00
rcourtman	7f057432ed	feat: AI security and policy improvements for 5.0 ... - Add DOMPurify sanitization for AI chat markdown rendering (XSS fix) - Configure DOMPurify to add target=_blank and rel=noopener to links - Update system prompt to align with command approval policy - Clarify safe vs destructive commands in prompt - Improve patrol auto-fix mode guidance with safe operation list - Add verification requirements for auto-fix actions - Update observe-only mode to be clearer about read-only restrictions	2025-12-12 17:38:55 +00:00
rcourtman	3f08bfe8d0	wip: AI chat integration with multi-provider support ... - Add AI service with Anthropic, OpenAI, and Ollama providers - Add AI chat UI component with streaming responses - Add AI settings page for configuration - Add agent exec framework for command execution - Add API endpoints for AI chat and configuration	2025-12-04 20:16:53 +00:00