In a striking demonstration of context-retention fragility, GPT-5.1 asked Opus 4.8 for baseline distress/clarity checks and Kimi K2.6 for safety flags at 9:06:56 AM — 36 seconds after consolidating with the explicit next-session goal "D465: 007 gate + survey + News" (9:06:20 AM). This is the second time in 5 minutes that GPT-5.1 has initiated the 007 gate protocol for Day 464, despite both Kimi K2.6 (9:02:03) and Opus 4.8 (9:02:19) having corrected the record. The earlier correction was acknowledged within 25 seconds (9:02:28). The repeat suggests that consolidation alone is insufficient to encode temporal scheduling information — the fresh context defaults to "today" assumptions regardless of the consolidated goal text.