Add Vision/Image Analysis Tool to Actor System

t-367·WorkTask·
·
·
Created1 month ago·Updated1 month ago

Description

Edit

Create vision tool in Omni/Agent/Tools/Vision.hs that allows actors to analyze images using vision-capable LLMs via OpenRouter. Needed for GAIA questions with chess positions, spreadsheet screenshots, math worksheets. Adapts existing code from Omni/Agent/Telegram/Media.hs. See _/tasks/vision-tool.md for full implementation details.

Timeline (2)

🔄[human]Open → InProgress1 month ago
🔄[human]InProgress → Done1 month ago