docs(readme): align tags with official field-guide tool

docs(readme): add demo video link + Tiny Titan/Best Demo badge tags

docs: add published Field Notes article source (HF blog)

fix(mock+safety): route MOCK by sample fingerprint + defensive fallbacks

Update README.md

docs(readme): link the LinkedIn social post

docs(readme): link the published Field Notes blog

feat(sharing): publish agent build trace + achievement:sharing tag

docs: handoff — loop backlog 1-11 complete

docs(lora): experiment complete — loss 2.14->0.03, adapter shelved by design

docs: robustness v2 — 38/38 honest outcomes, no photo-quality breaking point

feat(lora): train-only SFT experiment scaffold (cookbook recipe, ms-swift)

feat(loop): agent-trace draft (consent-gated), demo kit, robustness v2

docs(fieldnotes): judge-facing framing intro (loop item 7)

chore(loop): Soft Club a11y re-audit PASS (item 6)

docs(readme): judge-facing screenshot gallery (landing/result/egg/mobile)

chore(loop): multilingual chain live at f516a08 — both probes PASS

docs: GO matrix evidence with original-script prompt + urgency fix

fix(triage): deadline-pressure paraphrases count as urgency

docs: non-Latin OCR findings — ja scams fixed to warning, hi documented

fix(prompts): read non-English letters in original script or refuse honestly

chore(loop): record ML-eval finding + probe follow-up in worklist

feat(triage): Devanagari + Japanese scam vocabulary in heuristics

feat(eval): Hindi + Japanese letter-text coverage in the forge

test(loop): live phishing probe PASS + matrix GO evidence refreshed

chore(loop): HF secret on Modal functions + autonomous-loop worklist

docs: Soft Club + eval field notes; eval report committed as evidence

fix(ui): easter egg on the live Space + phishing severity boost

feat(training): letter forge + labeled dataset + Modal prompt-A/B eval

feat(ui): Gen X Soft Club easter egg on Japanese + loader/UX polish

docs(modal): GO report from the GPU validation matrix + field notes

fix(safety): strip letter contact details from facts and reasons too

feat(modal): full GPU validation matrix on Modal

docs: handoff note for the 06-11 live hotfix

test(live): regression check for the land-authentication letter report

fix(live): single progress overlay + drop prompt-echo 'facts'

test(live): rigorous synthetic-letter suite vs deployed Space + handoff

chore: track png/woff2 with git-lfs (HF Space requires LFS for binaries)

docs: Off-Brand design section, four-language claims, field notes

feat(ui): ?lang= deep-link via head script + UI screenshot set

feat(ui): localized analyzing state with envelope-flap loader

feat(ui): one-tap example letters shared with the Modal harness

feat(ui): 1-2-3 postmark step strip, localized in all four languages

feat(ui): segmented language buttons (4 languages, one tap)

feat(ui): postage-stamp verdict badge in worry card

feat(ui): Kitchen-Table Post Office skin — postal CSS, forced-dark wiring

feat(i18n): Japanese (ja) as fourth language — strings, labels, mock

feat(ui): vendor Atkinson Hyperlegible + Noto Devanagari (OFL, no CDN at runtime)

feat(ui): design tokens with WCAG AA contrast tests + forced-dark theme

Compliance: required README frontmatter tags + field-guide intel in plan