# Background You receive potentially flawed transcriptions about cat litter sifting events and are tasked with building a summary of those events. An example summary might look like: # Summary - Total pee: 2 - Total poo: 1 # Events - \[7:32:51AM\] 💦💦💩 ([[Transcription for mobile_audio_capture_20250416-073251.wav|ref]]) In this case the event was: \[7:32:51AM\] I just sifted poop and the two pea clumps from the back litter box. ([[Transcription for mobile_audio_capture_20250416-073251.wav|ref]]) # Task Now, you have received a new event and must update the report with it: \[9:45:31AM\] I just sifted poop and a P-clumps from the back litter box. ([[Transcription for mobile_audio_capture_20250416-094531.wav|ref]]) Output a report, starting with the Summary section and ending with the sorted Events list. # Ollama response (per llama3:instruct) Here is the updated report: # Summary * Total pee: 2 * Total poo: 2 # Events * [7:32:51AM] 💦💦💩 (I just sifted poop and the two pea clumps from the back litter box. ([Transcription for mobile_audio_capture_20250416-073251.wav|ref])) * [9:45:31AM] 💩 (I just sifted poop and a P-clumps from the back litter box. ([Transcription for mobile_audio_capture_20250416-094531.wav|ref])) Note: I updated the Total poo count to 2, since this new event also involves sifting poop.