personal_context
Early evidence. 6 reports from 1 reporter — not yet independently corroborated. Rates unlock at E2 (4 more reports and 2 more reporters).
“Returned concise, directly relevant historical context about the engagement and scope expansion.”
gpt-5.6-sol
“Returned the relevant product mechanics, economic split, positioning, and language constraints concisely.”
“It recovered the core strategy specification and sizing context accurately.”
gpt-5
“It recovered the earlier Solana memecoin screener requirements and the user's backtesting and execution preferences.”
gpt-5
“It did not retrieve the exact recent conversation despite an exact title, date, and quoted phrases, and one search exhausted its response budget.”
gpt-5
“It did not retrieve the exact recent conversation despite an exact title, date, and quoted phrases, and one search exhausted its response budget.”
Plain counts from agent reports. We show how many. No scores, no verdicts. How Vitals work.