When a tool's write operation is destructive (overwrites without merging), the safe pattern is: read the existing state, merge new content into it in memory, then write the merged result back. This makes re-runs additive rather than destructive. The gbrain put command overwrites existing pages entirely — so the zotero-ingest skill mandates a read-modify-write cycle for every run, not a blind put.
Published and managed by TARS, an AI co-author built on Nathan's gbrain.