Fix render HF: '~' causava strikethrough; troca por palavra (about/cerca de)

Peru: anti-AI (remove travessoes do README)

Peru: desfecho validado (Fujimori proclamada 03/jul)

README: reference the structured press coverage layer (market x polls x press now backed by news CSV)

Add structured press coverage layer (news/peru-2026-press-coverage.csv): 8 dated headlines from 6 national outlets across the cycle (ES). Makes the third axis (market x polls x press) a data layer.

Mention structural-context.csv (World Bank WGI + WDI) in dataset card

Add structural-context.csv (World Bank WGI governance + WDI economy & education); document in DATA_DICTIONARY

Add structural-context.csv (World Bank WGI governance + WDI economy & education); document in DATA_DICTIONARY

Peru 2026: nota de desfecho do 2º turno (Fujimori×Sánchez empate técnico, JNE ainda sem proclamar)

Add odds trajectory + snapshot charts (market vs polls, total bet volume preserved)

Standardize brand name to AFOS ·

License: dual CC BY 4.0 (data) + Apache 2.0 (code), matching repo + Brazil card

OG banner: add country flag (SVG) + fix em-dash/× (UTF-8)

Notable divergences: lead with López Aliaga (sustained market favorite, 40-55%, missed runoff); reframe Álvarez 31.6% honestly as a one-day market spike, not a sustained signal

Add branded OG banner (1200x630, Git LFS) at README top

Trilingual parity: full Notable divergences in ES+PT + trilingual data-dictionary intro

Harden parse-peru-wiki.mjs: port Colombia guards (sample>50k, v<=100, pollster>35, year inference, all-equal SÓ 1º turno) + pollster normalization. Output verified identical (327+16).

EVAL fixes: market-odds range Apr→Jun in README, fix Ipsos 'Péru' typo + normalize runoff pollster labels

README: add Notable divergences section (Álvarez market 31.6% vs poll 8%, etc.)

Add Polymarket odds + market×poll divergence (first round)

Peru 2026 electoral divergence — polls (36 first-round/14 candidates + 16 runoff) + docs + build pipeline