corpus update: README.md

README.md

@@ -17,7 +17,7 @@ size_categories:
 
 # Zambo Agent Execution Receipts
 
-Zambo is the verifiable work layer for AI: 152 tools behind one MCP endpoint, continuity that carries sessions, memory, and work across AIs, and a verifiable AER-1 receipt for every job, so "done" is something you can check, not something you have to trust.
+Zambo is the verifiable work layer for AI: 138+ tools behind one MCP endpoint, continuity that carries sessions, memory, and work across AIs, and a verifiable AER-1 receipt for every job, so "done" is something you can check, not something you have to trust.
 
 A growing public corpus of real AI tool-call execution receipts minted on
 [Zambo](https://zambo.dev). Every row
@@ -46,11 +46,11 @@ use looks like at scale.
 
 ## Dataset snapshot
 
-- Rows: 219
-- Unique receipt URLs: 219
+- Rows: 220
+- Unique receipt URLs: 220
 - Tools covered: 10 (capability_search, credithunt, ghost_audit_report, ghost_audit_site, ghost_audit_status, leadsignal, live_price, prompt_shield, provibe_audit, zambo_universal)
-- Date range: 2026-09-16 to 2026-10-16
-- Last updated: 2026-10-16
+- Date range: 2026-09-16 to 2026-10-17
+- Last updated: 2026-10-17
 
 ## Schema