diff --git a/src/content/statements/2026-08-11-weekly.md b/src/content/statements/2026-08-11-weekly.md
new file mode 100644
index 0000000..1d0cab1
--- /dev/null
+++ b/src/content/statements/2026-08-11-weekly.md
@@ -0,0 +1,39 @@
+---
+schema_version: 1
+type: weekly
+draft: false
+title: Week of 2026-08-05
+summary: ogsfrompoly weekly statement — 452 alerts, outcomes pending (0 markets resolved
+  in this period).
+period_start: '2026-08-05'
+period_end: '2026-08-11'
+bankroll_usd: 10000.0
+alert_count: 452
+hit_rate: 0.0
+resolved_count: 0
+hypothetical_pnl_usd: 491.15
+categories:
+- macro-finance
+top_wallets:
+- truncated_id: wallet_a637
+  category: macro-finance
+  hypothetical_pnl_usd: 47.999999999999986
+- truncated_id: wallet_300d
+  category: macro-finance
+  hypothetical_pnl_usd: 3.35
+- truncated_id: wallet_4f1a
+  category: macro-finance
+  hypothetical_pnl_usd: 5.800000000000011
+---
+
+## Strategy track record (2026-08-05 → 2026-08-11)
+
+Generated from ogsfrompoly warehouse for the weekly window 2026-08-05 → 2026-08-11. 452 alerts emitted in this window across the LP-public category (macro-finance). Numbers in the frontmatter are mechanically derived.
+
+Top wallets are ranked by realized PnL magnitude. A wallet-level `hypothetical_pnl_usd`, when present, is the $100-notional alert-model PnL for that wallet in this same window; it is an annotation, not the ranking key.
+
+## Methodology reminder
+
+See [/methodology](/methodology) for the sign-randomization skill test and the disclosure policy. All wallet references on this page are truncated IDs; no on-chain lookup is possible from this page alone.
+
+The hit rate scores the markets that **resolved** during this period, whenever they were alerted. Statements published before 2026-07-31 scored the alerts *emitted* in the period instead, so the two series are **not comparable** and should not be read as one track record.