myron 05522edb1d Add llava vision: use Ollama llava:7b for all image analysis, Claude as fallback ...

- Add _ollama_vision_call() using Ollama /api/generate with images array
- handle_screenshot: try llava first (free, on-LAN), fall back to Claude on error;
  text-only snapshots now use ollama instead of groq
- handle_vision: same llava-first/Claude-fallback pattern; caller can force
  provider='ollama' or 'claude' explicitly; stored provider_used reflects actual
  provider that ran

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01X8tDRrQqgLjqXebMCBNcP3

2026-07-01 04:05:37 -05:00

..

cron-jarvis.txt

…

jarvis-agent.service

…

jarvis-arc.service

Fix Arc Reactor restart: use bash for source, add systemd service, fix deploy

2026-07-01 03:56:11 -05:00

jarvis-deploy.sh

Wire Ollama properly: fix IP, upgrade to llama3.1:8b, use for guardian/sitrep

2026-07-01 04:02:30 -05:00

jarvis-watchdog.sh

Fix 8 code-review findings: security + reliability

2026-05-25 14:27:27 +00:00

reactor.py

Add llava vision: use Ollama llava:7b for all image analysis, Claude as fallback

2026-07-01 04:05:37 -05:00