dbddf17e3b6e660af21c55ff5789cb0796537d87
9 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 Kwaku Danso
|
b34715f152 |
fix(tier2-g): warmer Gate copy, generic example, plain-language advanced controls
Round-three pass on the Gate card based on the latest UX feedback. - Tagline: "Keeps your guest list yours" -> "Only your guests get in" - Walk-through example now uses a neutral placeholder name (Sam) instead of "Aunty Patience". GuestGuard's audience is wedding couples, corporate planners, and party hosts as much as family organisers, and Sam is at home in any of those contexts. - The four-step "How does the Gate work?" expander was lightly rewritten for warmth and to remove em-dashes; the mechanism it describes is unchanged. - Advanced strictness controls now open with a real explanation rather than a one-liner about "band thresholds". The intro maps the three sliders to the three reactions (watch / flag / refuse), explains what the 0-100 numbers mean, and reassures hosts that the presets above are still fine for almost everyone. Each slider also gets a one-line caption tying its level dot (green / amber / red) to the user-visible effect. - The trusted-networks "What this is and isn't" panel had its quotes unwound into natural sentences; the empty-state copy was rewritten in the second person; "No decisions to review yet" reworded. - Pass through every user-visible string and replaced em-dashes with periods, commas, or natural rephrasing. Em-dashes only remain in code comments now (developer-facing, not on screen). No behaviour changes; no backend changes; no API changes. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> |
||
|
Kwaku Danso
|
dee3d738ac |
fix(tier2-g): explain the Gate clearly, kill the "everyone outside = fraud" misconception
The previous pass made the labels friendlier but didn't actually explain what the Gate does. A real user (correctly) read "trusted networks" as "any guest outside my IP is treated as fraud," which is the opposite of what's happening — and missed the entire value prop: stopping forwarded invitation links from being used by uninvited people. That's the central pain GuestGuard solves; the Gate page has to lead with it, not bury it under controls. What changed on the Gate card: - Big value-prop block at the top in plain language: "Every guest gets their own personal invitation link. The Gate watches each link and stops forwarded or shared invitations from being used by people who weren't on your list." Also explicit: "Real invited guests don't notice it. You don't need to set anything up — the Gate is on by default with sensible settings." - New "How does the Gate work?" collapsible explainer with a 4-step walkthrough using Aunty Patience as the protagonist. Covers what signals the Gate looks at, when guests *do* get flagged, and the reassurance that normal day-to-day variation (Wi-Fi → mobile data, changing rooms) doesn't trigger anything. - Preset descriptions rewritten to talk about the actual pain instead of generic strictness levels. "Stops forwarded links from being used by people who weren't invited" lands much harder than "Recommended for most parties." - Trusted networks section opens with an explicit "What this is and isn't" panel that directly addresses the misconception: adding a network here is a *speed-up* for guests on your Wi-Fi, NOT a whitelist that classifies everyone else as suspicious. Empty state reads "No trusted networks — and that's fine. The Gate is doing its job" so a host doesn't feel they've missed a setup step. - Pill on the section header tags it "Optional · most hosts don't need this" to drop its prominence. The button copy went from "Use my current network" (which read like "approve myself") to "Trust the network I'm on right now." No backend changes. Internal Go names (FraudThresholds, /security/* endpoints) untouched — never user-visible, renaming would churn for zero end-user benefit. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> |
||
|
Kwaku Danso
|
3629ab8c79 |
fix(tier2-g): rebrand Security → Gate with event-planner language
The original Block G UI was correct but felt like a firewall config panel.
The audience for GuestGuard is wedding couples, party hosts, and event
planners — not network admins. Three concrete problems:
1. "Fraud detection" / "Risk score" frames every guest as a suspect.
"Gate" is the metaphor every host already uses ("gate the guest
list", "doorman"). It's the same idea, friendlier.
2. Three threshold sliders (0–100) are abstract. Hosts can't reason
about "block at 85" without context. Replaced by four presets —
Relaxed / Balanced / Strict / Very strict — each with a sentence
that explains the kind of event it fits. The sliders survive under
an Advanced disclosure for the rare power user.
3. CIDR notation is gibberish to a layperson. "Trusted networks" is
what they're actually doing. A "Use my current network" button
detects the host's apparent IP, widens IPv4 to /24 (typical home
block), and pre-fills the form. Manual entry stays for cases
where the host knows what they're doing.
Plus two leaks fixed on the guest-facing RSVP page: removed the
"Risk score 72 · high" line from both the confirmation and the
blocked-attempt cards. Guests should never see internal scoring
detail — the blocked message now reads "Something about this
attempt looked off" instead of "suspicious access attempt".
Backend
- New GET /me/public-ip (authed) — echoes the caller's apparent IP
so the frontend's auto-detect button doesn't need a third-party
ipify call
Frontend
- New components/GateCard.vue with preset cards + advanced disclosure
+ trusted-networks UX with auto-detect
- Removed components/SecurityCard.vue
- Event tab nav: Security → Gate. Hash-link alias preserves any
/events/<id>#security bookmarks
- RSVP page: leaked risk-score lines removed; blocked-attempt
message reworded for non-technical guests
Internal storage / API names stay (FraudThresholds, fraud_v2,
/security/* endpoints) — they're never user-visible and renaming
them would be a breaking change for no end-user benefit.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
|
||
|
Kwaku Danso
|
b873012191 |
feat(tier2): smarter fraud detection — Block G
Per-event fraud tuning. Hosts can now dial the medium / high / block
boundaries, allowlist trusted networks, and feed verdicts back on
flagged accesses — the seed corpus for a future ML model.
Schema (migration 0011)
- events.fraud_{medium,high,block}_threshold default 30/60/85 so
existing events behave identically until a host changes them
- access_logs.geo_{country,city,lat,lon} for future enrichment
- fraud_feedback table — verdict ('legitimate' | 'suspicious') + note,
PK on access_log_id so re-mark is an upsert
- event_allowlists table — (event_id, ip_cidr) primary key, inet column
so containment checks use the native >>= operator (indexed lookup)
Domain
- FraudThresholds with Valid() + Band() helpers; Default trio echoed
through GET responses so the frontend doesn't duplicate constants
- ParseAllowlistCIDR accepts bare IPs (auto-widens to /32 or /128) and
canonicalises the output (203.0.113.42 → 203.0.113.42/32)
- Event.Thresholds() falls back to defaults if columns weren't
populated yet, so the API never wedges every score into "low"
Storage
- AllowlistRepo: List / Add / Remove + Matches() — the latter pushes
CIDR containment into Postgres rather than streaming rows back
- FeedbackRepo: Record (upserts) + ListForEvent (joined through guests)
- EventRepo.GetThresholds + UpdateThresholds, plus the threshold
columns baked into scanEvent so every event load carries them
- AccessLogRepo.BelongsToEvent — stops a hostile editor on event A
from marking event B's access logs
API
- GET/PUT /events/{id}/security/thresholds (viewer/editor)
- GET/POST/DELETE /events/{id}/security/allowlist
- POST /events/{id}/access-logs/{log_id}/feedback (editor)
- GET /events/{id}/security/feedback
- RSVP scoring path: allowlist short-circuit fires before the fraud
engine; the engine's score is then re-banded against the event's
thresholds (engine.Risk becomes advisory — API is the source of
truth for "what counts as block here")
- CORS Allow-Methods already includes PUT (Block D fix)
Fraud engine
- Single-signal cap: it now takes ≥2 sub-scores of ≥70 to push the
final into HIGH. Fixes the well-known "second visit with a slightly
shifted fingerprint scores 60+" false positive
- Engine band remains advisory; API re-bands using per-event
thresholds before deciding to block
Frontend
- SecurityCard.vue: visual band ribbon (proportional to thresholds),
three sliders with mutual clamping so dragging medium past high
pushes high (not an invalid ordering), reset-to-defaults button,
CIDR allowlist with inline add + per-row remove, verdict-history
inbox. Toast feedback on save/add/remove
- "Security" tab added to the event-detail tab nav (5th tab,
right of Analytics)
- Viewer role hides write affordances; server enforces too
Tests
- Domain: ThresholdsBand, ThresholdsValid, ParseAllowlistCIDR (bare
IP widening + traversal/typo rejection), FraudFeedbackValid
- Integration: thresholds round-trip + invalid ordering rejection,
allowlist CRUD + duplicate 409 + invalid CIDR 400 + IP auto-widen,
feedback record + upsert + cross-tenant 404 + invalid verdict 400,
viewer can read / editor can write / outsider gets 404
- Full integration suite green (315.8s, all 36 top-level tests pass)
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
|
||
|
Kwaku Danso
|
e5b187c575 |
feat(tier2): event branding + UX polish — Block D
Backend
- Migration 0010 adds event_branding (one row per event; all fields
nullable so a brand-new event renders with defaults)
- BrandingRepo with COALESCE/NULLIF upsert semantics: nil pointer
preserves the existing value, "" clears the field to NULL
- internal/uploads package: ImageStore interface + LocalFSStore (dev),
pure-stdlib decode + re-encode that strips EXIF and rejects anything
that isn't valid JPEG/PNG. Size cap 2 MB, random 16-byte filenames
- GET /events/{id}/branding (viewer+) returns the row plus the
AllowedFonts list so the frontend picker stays in sync
- PUT /events/{id}/branding (editor+) validates hex colours, font
allowlist, and refuses image URLs whose path doesn't start with
/uploads/ (blocks arbitrary-origin <img> smuggling on guest pages)
- POST /uploads/image (authed) → fresh CDN URL; GET /uploads/{file}
serves with year-long cache (immutable random names)
- GET /access/{token} now embeds the host's branding so the RSVP page
can render in their colours/font with their logo + cover
- docker-compose mounts a named volume for uploads
- Custom-domain sub-block deferred to Tier 3 per the plan
Frontend
- BrandingCard.vue: colour pickers, font dropdown, logo + cover upload
with progressive disclosure, live preview pane that re-renders on
every keystroke
- RSVP page applies branding via CSS vars at the section root, so
primary colour theme + font cascade through every child card. Cover
image renders as a banner above the form; logo lands in the header
- Submit button background switches to var(--brand-primary) when set
- Mounted on the event detail page below the guests block
Plus the small UX fixes from the e2e walkthrough:
- Nav: dropped the top-level "Events" link; the logo doubles as the
home affordance (→ /dashboard when signed in, → / otherwise). Account
+ Billing + Sign out live under a profile dropdown (avatar with
initials, opens on click, closes on outside-click / Esc / route nav)
- Renamed "Back to dashboard" → "Back to events" across event detail,
billing, account, and new-event pages
Tests
- TestBrandingGetReturnsDefaults / TestBrandingPutPersists /
TestBrandingPutRejectsBadInputs / TestUploadAndServeImage /
TestUploadRejectsNonImage — all pass
- Domain tests for IsValidHexColor + IsAllowedFont
- Full integration suite green (176s)
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
|
||
|
Kwaku Danso
|
9842bd4f45 |
feat(tier2): host analytics — Block E
GET /events/{id}/analytics renders a viewer+ dashboard summary:
overview tiles, 30-day response sparkline, invited→opened→responded
funnel, time-to-respond histogram, plus-ones distribution, channel
attribution (utm_source), and stale-guest follow-up list. A
matching /analytics/export.csv hand-offs a flat per-guest table for
Excel + Numbers.
Backend
- Migration 0009 adds tokens.utm_source for source attribution
- AnalyticsRepo with six aggregation queries — all event-scoped, all
returning canonical-ordered series so empty buckets still render
- Redis 60s cache in the handler, keyed by (event_id, days). Cache
miss path returns X-Cache: miss; hit returns the JSON straight
from Redis without re-querying Postgres
- Time-to-respond uses (rsvp.submitted_at - token.created_at) as
the latency signal (no separate "invitation sent" timestamp yet —
Block F will add one)
Frontend
- AnalyticsCard.vue: inline SVG sparkline + Tailwind bar charts.
No chart.js dependency; the bundle stays lean
- Stale-guests list with opened-but-no-response highlighted
- Export CSV button issues an authed fetch + blob download
Tests
- TestAnalyticsAggregations seeds 5 guests with a known mix and
asserts every count (overview/funnel/plus-ones/time-to-respond/
stale) matches expected
- TestAnalyticsCSVExport: header row + per-guest rows parse cleanly
- TestAnalyticsAuthzMatrix: viewer 200, outsider 404 on both endpoints
- Full integration suite passes (109.9s)
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
|
||
|
Kwaku Danso
|
3973e4058d |
feat(tier2): multi-host / collaborators — Block C
Events can now have multiple users with distinct roles:
owner — manage collaborators, delete event, full access
editor — manage guests, tokens, CSV import, patch event
viewer — read-only access to everything
Schema (migration 0008)
- collaborator_role ENUM + event_collaborators + collaborator_invites
- Backfill: every existing events.host_id becomes an owner row
- EventRepo.Create seeds the owner row in the same transaction so
no future event can exist without one
Authz
- New requireRole(eventID, userID, minRole) helper. Non-members 404;
insufficient role 403. Replaces requireEventOwner across every
shared-role handler (events.get/update, guests CRUD, tokens issue/
rotate/bulk, csv preview/commit/template, activity, ws-ticket)
- events.delete + collaborator management stay owner-only
- GET /events lists every event the user has any role on
- /events/{id} response now embeds your_role for UI branching
Collaborator endpoints
- GET /events/{id}/collaborators (viewer+)
- POST /events/{id}/collaborators (owner) — sends invite email
- PATCH /events/{id}/collaborators/{user_id} (owner) — role change
- DELETE /events/{id}/collaborators/{user_id} (owner) — refuses last owner
- DELETE /events/{id}/collaborators/pending (owner) — cancel invite
- GET /invites/{token} (public) — preview summary
- POST /invites/{token}/accept (authed) — atomic accept
Invitations
- SHA-256 hashed in DB; raw value only lives in the email link
- 7-day TTL, single-use, email-bound (caller's email must match)
- New SendCollaboratorInvite on auth.EmailSender + Resend/SMTP/SES
senders + log stub; collaborator_invite.html/txt branded template
Frontend
- TeamCard.vue on the event detail page: lists collaborators with
inline role-change + remove, pending-invites with cancel, invite
modal (email + role). Owner-only actions hidden for editors/viewers
- /invites/[token] accept page: shows invite summary, prompts signup
or sign-in with pre-filled email, refuses mismatched accounts
Tests (all 6 pass on the existing testcontainers harness)
- backfill: legacy host gets owner role
- role enforcement: viewer can read, editor can write guests but not
delete/manage team, non-member 404s everywhere
- last-owner removal refused (400)
- shared events show up in collaborator's /events list
- invite flow: create → preview → accept → role granted → replay 410
- email mismatch on accept returns 403
- expired invite returns 410
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
|
||
|
Kwaku Danso
|
6803d700b4 |
feat(tier2): calendar integration — Block B
After confirming attendance (or revisiting an already-attending RSVP),
guests can add the event to Google, Outlook, Apple, or any iCalendar
client with one click.
- internal/calendar package builds an RFC 5545 VCALENDAR plus the three
provider deep-links from a narrow Event projection. Times are emitted
as Z-suffixed UTC; the UID is deterministic (event-uuid@guestguard)
so re-downloads update the same calendar entry instead of duplicating
- GET /access/{token}/calendar.ics returns the .ics with a slugified
filename in Content-Disposition; same rate-limit class as /access
- GET /access/{token} response now embeds google/outlook/yahoo/ics URLs
so the frontend doesn't re-encode per provider
- AddToCalendar.vue renders the four buttons; shown only when the
guest is attending (declined/maybe don't need a calendar entry)
- Unit tests cover ics field presence, RFC 5545 escaping (comma,
semicolon), CRLF line endings, explicit-end handling, omitted
optionals, provider URL shape, filename slugification
- Integration: ics download returns valid VCALENDAR with the event
UID + name; unknown token returns 404; /access embeds the links
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
|
||
|
Kwaku Danso
|
59b8781659 |
feat: ship Tier 1 — auth, authz, rate limits, real notifications, CSV import, billing, backups/DR, privacy
Closes every block in docs/TIER1_PLAN.md from the Claude-scope side. The
homelab / cloud setup steps (SES verification, restore drill, lawyer-
drafted ToS) remain operator-owned but are unblocked.
Block A — Authentication
- Migration 0003: password_hash, email_verified, email_verification_tokens,
password_reset_tokens, refresh_tokens (with replaced_by family chain).
- Bcrypt hasher, HS256 JWT signer, single-use refresh tokens with rotation
+ replay-detection (revokes the family on reuse).
- /auth/signup, /login, /refresh, /logout, /verify-email,
/forgot-password, /reset-password — enumeration-safe.
- requireAuth middleware + GET /me.
- Frontend useAuth/useApi with auto-refresh-on-401, login/signup/verify/
forgot/reset pages, route-guard middleware.
Block B — Authorisation
- EventRepo.GetForHost; Update/Delete scoped by host_id.
- All host routes behind requireAuth + ownership; cross-tenant returns
404 (no enumeration). ?host_id removed.
- WS auth via short-lived single-use tickets (POST /auth/ws-ticket).
- Tests: TestCrossTenantIsolation — 9 probes.
Block C — Rate limiting
- Redis sliding-window via Lua (atomic ZADD+ZCARD+PEXPIRE).
- Per-route limits matching the plan (signup IP, login IP+email, RSVP/
access by token, events/guests/tokens by user_id).
- 429 with Retry-After header and JSON body.
- Auth lockout: 5 failed logins → account locked, only password reset
clears it.
- Frontend: useErrMessage normalises 429 + locked messaging.
Block D — Real notifications
- Migration 0004: provider_message_id, bounce_type, complained columns
+ unsubscribes (CITEXT) suppression table.
- Branded HTML + plaintext templates for verification, reset, invitation,
confirmation, reminder. Per-page templates avoid html/template's
contextual-escape collisions.
- Senders: SESv2, Twilio (SMS), SMTP (Mailpit-friendly), Resend HTTP.
- PickEmailSender priority Resend > SMTP > SES > Log — system boots
cleanly in dev with Mailpit; production flips one env var.
- Webhook endpoints (Twilio status + SES SNS) — bounces add to suppression;
signature verification stubbed pending creds.
- Auto-send: POST /tokens publishes invitation.send; notifier renders +
delivers via the configured backend; suppression list honoured.
- Bulk + per-row invitation flow: POST /events/{id}/guests/invitations/bulk
returns per-guest tokens so phone-only guests can be SMS'd manually.
- Unsubscribe: signed HMAC token (no TTL) + /unsubscribe/[token] page.
- WhatsApp Option A+: wa.me click-to-chat wizard with per-guest progress
tracking, isLikelyE164 validation, edit-from-wizard.
- Token rotate (POST /tokens/rotate) invalidates the old URL — used by
the regenerate-link flow.
- Mailpit added to docker-compose for dev inbox.
Block E — CSV import
- Streaming parser: tolerant header detection, UTF-8 BOM + UTF-16 LE/BE
decoding, row-level validation, 5,000-row cap.
- Strict E.164 phone validation with helpful error message.
- POST /preview + /import + GET /template; preview UI on event page;
atomic per-batch with dedup on existing emails.
Phone capture across UI
- PhoneInput component: country picker (~50 ISO codes) + national input +
live E.164 preview + inline length validation.
- Used in Add Guest and Edit Guest modals. Smart paste-handling extracts
country code from full E.164 strings.
Block F — Billing (Stripe)
- Migration 0005: subscriptions table (user_id → tier/status/period_end +
Stripe customer/sub ids). Partial unique index keeps one granting sub
per user.
- internal/billing: Tier + Limits model (Free 1/50, Pro 10/1000, Business
∞/5000), Stripe SDK wrapper with IgnoreAPIVersionMismatch for newer
account API versions.
- /billing/checkout-session, /billing/portal, /billing/status,
/webhooks/stripe (signature-verified, lifecycle events).
- Tier enforcement: 402 on POST /events, /guests, /import with
{error, reason, tier, used, limit, upgrade_url} body.
- Frontend: useBilling composable, /dashboard/billing page (current plan,
usage bars, tier cards), global UpgradeModal triggered by useApi's
402 interceptor.
- Customer portal kept for self-service cancel/payment-method changes.
Block G — Backups & DR (application side)
- Every migration has a tested .down.sql.
- TestMigrationRoundtrip applies all ups → all downs → all ups against a
fresh container; catches asymmetric down migrations.
- cmd/restore-verify: 28-check post-restore invariant tool (schema
presence, no orphans across 10 FK relationships, email uniqueness,
single-active subscription, row-count snapshot).
- docs/RUNBOOK_RESTORE.md: 9-step restore procedure with RTO/RPO
targets, drill instructions, rollback path.
Block H — Privacy compliance (application side)
- Migration 0006: deleted_at + terms_accepted_at + privacy_policy_accepted_at
on users. Partial index on email for live-only uniqueness.
- GET /me/data-export — synchronous JSON dump (user, events, guests,
tokens, rsvps, access_logs, notifications).
- DELETE /me — soft-delete with PII scrub + refresh-token revocation;
re-signup with same email works.
- POST /me/accept-terms — idempotent consent recording.
- Frontend /privacy + /terms placeholder pages with substantive (pending
legal review) copy; footer links; signup terms checkbox; TermsGateModal
for accounts created before the rollout; export + delete buttons on
/dashboard/billing.
Tests
- All migrations verified up/down/up.
- Integration suite: TestE2EHappyPath, TestAuthFlow, TestCrossTenantIsolation,
TestRateLimitSignup, TestLoginLockout, TestUnsubscribeFlow,
TestSESBounceWebhook, TestTwilioStatusWebhook, TestCsvImportFlow,
TestCsvImportAtomicRollback, TestBulkIssueInvitations, TestBulkIssueExplicitSubset,
TestTokenIssuePublishesInvitation, TestTokenIssueWithoutGuestEmailSkipsInvitation,
TestGuestUpdate, TestGuestDelete, TestTokenRotate, TestSMTPSenderAgainstMailpit,
TestFreeTierEventLimit, TestFreeTierGuestLimit, TestBusinessTierBypassesLimits,
TestDataExport, TestDeleteMe, TestAcceptTerms, TestMigrationRoundtrip.
Full suite runs in ~120s against real Postgres + NATS + Redis + Mailpit.
- Unit suite green across internal/auth, internal/csvimport,
internal/notification, internal/ratelimit, internal/domain.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
|