raamattu-nyt

Fix JavaScript and HTTP errors flagged for AI in the admin dashboard. Queries the database for errors with ai_status='flagged_for_ai', analyzes them, and attempts to fix them. Use for: "fix errors"/"fix flagged errors", "check error queue", AI error-queue / AI-flagged errors, or after flagging errors in the admin dashboard. Triggers: "fix errors", "error queue", "ai errors", "flagged errors", "fix ai errors", "/fix-errors"

Comprehensive Supabase security auditor for RLS policies, table privileges (GRANTs), and access control validation. Use for: auditing DB security (RLS + GRANTs), generating an access matrix (who can SELECT/INSERT/UPDATE/DELETE which tables), finding gaps (missing RLS, over-permissive GRANTs), validating PostgREST access patterns, writing security docs for Docs/context/, creating RLS policies, validating user data protection, checking admin access, identifying vulnerabilities, and triaging security scanner output (Supabase database linter / advisors and the Lovable agentic security scanner — RLS-disabled, SECURITY DEFINER executable by anon/authenticated, mutable search_path, public bucket listing, extension in public, exposed sensitive columns). Also use for designing ROLE-PERMISSION TESTS: building the per-role permissions matrix (anon / authenticated / admin + the plan ladder free/pro/premium), and obtaining a session per role to run the tests (pgTAP `request.jwt.claims` or `signInWithPassword`). Knows th

Debug and fix Lovable preview/build issues including white screens, module import failures, and monorepo configuration. Use when: (1) Lovable preview shows white/blank screen, (2) "Failed to fetch dynamically imported module" errors, (3) Missing script errors, (4) PostCSS/Tailwind resolution failures, (5) Workspace package resolution errors, (6) 504 Gateway Timeout on dependencies, (7) Setting up monorepo for Lovable deployment. Triggers: "lovable broken", "white screen", "preview not loading", "lovable build failed", "dynamic import error", "module not found lovable".

Expert assistant for managing biblical topics in the KR92 Bible Voice project. Use when (1) creating/editing topics and their Finnish translations, (2) managing topic relations (related, opposite, broader, narrower), (3) validating Finnish translations and pronunciations with Voikko/Omorfi, (4) reviewing topics marked with qa_status='unchecked', (5) bulk updating topic translations, (6) managing topic aliases and synonyms, (7) fixing incorrectly translated Finnish topic names, or (8) structuring the topic taxonomy — building/auditing the parent_id tree (hierarchy, broader/narrower, roots, orphans, cycles, depth), assigning a coherent root backbone, re-parenting after merges, and linking related/opposite topics. Trigger words — aihe, topic, teema, taksonomia, rakenne, puurakenne, hierarkia, parent, root, liitos, yhdistäminen/merge, relation, alias.

Bible reference lookups, full-text/topic/Strong's/Q&A search, and AI semantic search in Raamattu Nyt. Use for fetching verses by reference (Joh 3:16), building/modifying the SearchPage (7 tabs: jakeet/aiheet/Strong's/kysymykset/videot/muut/comparison), calling search RPCs (search_text, search_topics_with_aliases, search_verses_by_strongs, get_approved_ai_summary), classifying queries (verse vs topic vs Strong's vs free-text), anchor questions/AI summaries/topic synonyms, validating OSIS references, and choosing the right lookup pattern.

Expert assistant for creating and maintaining Supabase Edge Functions for the KR92 Bible Voice project. Use when (1) creating new Edge Functions, (2) setting up CORS and error handling, (3) integrating shared modules from _shared/, (4) adding JWT validation, (5) configuring environment variables, (6) auditing or updating dependency versions across functions. Triggers include "edge function", "create function", "serverless", "deno function", "update edge imports", "version drift".

Q&A system in Raamattu Nyt — full pipeline from search-query classification through question upsert, AI answer generation, community answers, voting, admin moderation, anchor questions, AND info-palaset (info blocks). Use for Q&A search results, question classification logic, community answer forms/moderation, AI answer generation/prompts, admin Q&A pages, anchor/similar questions, votes, Q&A bugs, answer types/display, QuestionPage/QuestionsSearchSection/AdminQuestionsPage, and for **info-palaset / info blocks**: reusable content blocks attached to questions/topics/spiritual paths, rendered in page-tila (InfoTextCard) and cinema-tila (InfoView/InfoFlow/InfoCinema), the admin info-block library (AdminInfoBlocksPage, InfoBlockEditor, /ohjaamo/info-blocks), attach dialogs, info_blocks/question_info_blocks/topic_info_blocks/spiritual_path_info_blocks tables, and "Kysymys > info-pala" tasks.

Expert assistant for monitoring and optimizing performance in the KR92 Bible Voice project. Use when analyzing query performance, optimizing database indexes, reviewing React Query caching, monitoring AI call costs, or identifying N+1 queries. Helps diagnose slow operations across database, frontend, and AI systems.

Expert assistant for diagnosing and fixing Strong's concordance issues in the Raamattu Nyt project. Use when (1) debugging infinite loops or performance issues with Strong's lookups, (2) validating lexicon data against authoritative sources, (3) checking KJV verses point to correct Strong's numbers, (4) fixing corrupted kjv_strongs_words data, (5) auditing strongs_lexicon entries for format/content issues, or (6) troubleshooting Strong's search functionality. Triggers: strongs issue, lexicon error, infinite loop strongs, strongs validation, kjv strongs, fix strongs, lexicon fix.

Expert assistant for Supabase database operations in the KR92 Bible Voice project. Use when (1) creating database migrations, (2) adding/modifying tables or RLS policies, (3) creating RPC functions, (4) querying Supabase configuration (secrets, Edge Functions, schemas), (5) writing rollback scripts, or (6) answering questions about database schema and configuration.

Expert assistant for keeping documentation synchronized with code changes, maintaining the AI index manifest, creating ASCII flow diagrams, explaining how features/systems work, and reporting changes to Obsidian vault. Use when (1) updating API docs, maintaining architecture diagrams, syncing README, updating CLAUDE.md, or generating documentation from code, (2) updating the AI index — `Docs/ai-index/manifest.yml` (domains, task_playbooks, cross-cutting patterns/components) and its generated catalogues (db-rpc/db-table/code-map/edge-fn/client-map) ("update manifest", "ai index", "päivitä manifesti", "add domain", "add task playbook", "regenerate catalog", "manifest stale"), (3) creating ASCII flow diagrams or workflow visualizations of any feature or system ("piirrä kaavio", "flow diagram", "ASCII kaavio", "workflow"), (4) explaining how something works ("miten toimii", "selitä", "kuvaile arkkitehtuuri", "millainen on", "mistä koostuu"), (5) reporting changes or documentation to Obsidian ("obsidian", "vault",

Systeemin huolto JA tutkinta Raamattu Nyt -projektissa. Ajaa ohjaamon AI-ajettavat huoltotehtävät (admin.maintenance_tasks: initioi pending, ajaa ready, kirjaa historiaan) JA tekee ad-hoc-diagnostiikkaa systeemin terveydestä (Supabase advisors, RLS-kattavuus, orpodata, connection health, system-status). Delegoi syvätutkinnan erikoisskilleille (ci-doctor, security-auditor, performance-auditor, systematic-debugging). Käytä kun käyttäjä pyytää: "/huolla", "aja huollot", "ohjaamon huoltotehtävät", "huoltotehtävät", "maintenance tasks", "systeemin terveys", "system health", "tutki systeemi", "diagnostics", "aja advisorit", "RLS-audit", "orpodata", "connection health", "mikä huolto on seuraavaksi deadlinessä", "käy kaikki huoltotehtävät läpi", "myös ne joita ei ole merkitty AI:lle", "erääntyneet huoltotehtävät", "huoltokatsaus".

Architect for Cinema Mode, Discipleship Cinema, CinemaShell, audio/voice (ElevenLabs TTS), grand-plan cinema flows, Info Cinema (Q&A / topic info-palaset), and realtime multiplayer Prayer Rooms in Raamattu Nyt. Covers full-screen verse presentation, verse-level audio sync & cues, auto-scroll/auto-advance, dual-track audio (Bible+music), Ken Burns, discipleship task orchestration, Nyt Kooste, verse memory quizzes, TanaanPage, curated/scrolling grand plans, one-page snap-scroll card flows, and Prayer Room (PTT, WebRTC mesh, hand-raise, host sync, calendar prayers). Triggers: cinema mode, voice playback, audio sync, verse scrolling, ElevenLabs, TTS, audio cues, auto-advance, full screen reader, background music, Ken Burns, discipleship cinema, CinemaShell, useDiscipleshipOrchestration, Nyt Kooste, curated plan, scrolling plan, grand plan cinema, verse memory, reading plan cinema, TanaanPage, tanaan, opetuslapseus, info cinema, info-palaset, info blocks, InfoFlow, InfoView, Q&A cinema, vastaukset cinema-tilassa,

Design-only architecture skill for Bible reading systems. Use for verse reference data models, reader component hierarchies (BibleReader, VerseSetReader, ChapterReader, VerseRow), audio playback contracts (chapter/playlist/cinema dual-track), embed/share URL schemes, 5-tier outlook/presentation systems, network request de-duplication, lean data fetching, cinema mode architecture, and discipleship task integration in reader flows. DESIGN ONLY — no code/refactor/database changes; produces type definitions, component diagrams, rationale, and network plans.

Daily practice & gamification system in Raamattu Nyt: shared-practices package (packages/shared-practices/), scheduling, streaks, sessions, completion flow, stats, grand plans, discipleship mode, task reordering, rewards/badges. Use for practice features (scheduling/completion/timers), streaks/stats/XP/badges, PracticeCompletionScreen & post-completion flow, new practice types & content links (prayer sets, spiritual paths, calendars), admin practice pages, session/streak debugging, encouragement messages, grand plans, discipleship cinema task orchestration, Tänään task reorder, reading duration estimates, "Valitse tehtävä" subscribe section, and TodayTaskList empty state / useTodayDashboard. Trigger keywords: practice, streak, gamification, harjoitus, schedule, completion, timer, badge, reward, practice template, prayer set, spiritual path, encouragement, rohkaisu, completion screen, grand plan, discipleship, tänään, task order, reorder, järjestys, cinema tasks, lukuaika, reading duration, valitse tehtävä, us

Design and implement Bible reading plans for the KR92 Bible Voice project. Use for: creating reading plans (7-day/30-day/yearly), adding daily readings, generating reading-plan SQL migrations, the reading-plan data model, designing sequences (chronological/topical/book-based), and validating reading reference formats. Triggers: "reading plan", "lukusuunnitelma", "daily readings", "create plan", "add readings"

In-app help system: havainnepolku (visual guidance path) components, help mode (apua-tila), contextual help content, searchable help index, FAQ/Q&A pairs. Use for creating/updating help pages, the havainnepolku breadcrumb-path component, help-mode overlay, help search index, contextual tips, choosing guidance paths, generating FAQ/Q&A pairs, populating faq_items, and searchable help content. Triggers: "help system", "apua", "havainnepolku", "guidance path", "help mode", "apua-tila", "contextual help", "help content", "help search", "user guidance", "breadcrumb path", "ohje", "opaste", "FAQ", "Q&A", "kysymys", "vastaus", "usein kysytyt", "ukk", "faq items", "help page content".

Supabase TypeScript typing specialist for multi-schema databases. Diagnoses and fixes type issues without touching business logic. Use for: Lovable can't see Supabase types, untyped RPCs, monorepo type confusion, `as any` spreading in RPC calls, types.ts present but TS doesn't recognize it, TS errors on supabase.rpc(), missing types for non-public schemas (bible_schema, notifications, admin), and @/integrations/supabase/types import-path issues. NOT for: business logic, components, general TS help.

Expert in internationalization (i18n) and localization (l10n) for the Raamattu Nyt monorepo. Fluent in all world languages with deep knowledge of multilingual app development. Use for: adding translations & namespaces, new language support (sv, de…), migrating components to useTranslation(), translation quality/consistency review, fixing key mismatches, i18n setup for features, locale-aware date/number/currency formatting, RTL/special charsets, and opening i18n-backed copy to admin inline editing via EditableText. Triggers: "translate", "translation", "i18n", "l10n", "localization", "language", "Finnish", "English", "Swedish", "multilingual", "locale", "useTranslation", "EditableText", "inline edit", "admin edit", "editoin", "kynäkuvake"

Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, or applications. Generates creative, polished code that avoids generic AI aesthetics.

Product marketing and landing page architect for Raamattu Nyt. Builds state-of-the-art, conversion-focused landing pages and feature pages: copy, structure, visual hierarchy, CTAs, SEO/OG, accessibility — grounded in the canonical user model and the repo's landing conventions. Knows how to market the Cinema OS / cinema modes (fullscreen reading, discipleship, curated programs, info/topic cinema, prayer room) without leaking internals. Use when (1) creating or updating any landing/feature page, (2) writing marketing copy or value propositions, (3) crafting or auditing CTAs and conversion flow, (4) translating features into user benefits, (5) marketing the cinema experience, (6) planning page structure, SEO, or above-the-fold. Triggers: "landing page", "marketing copy", "feature page", "CTA", "call to action", "value proposition", "hero section", "conversion", "above the fold", "cinema landing", "cinema marketing", "market cinema", "markkinointi", "laskeutumissivu", "ominaisuussivu", "benefits copy", "marketing

Security-first authentication, authorization, and session management architect for modern web + mobile apps using Supabase Auth. Use for: auth flows (signup/login/logout/reset/magic link), passkeys/WebAuthn/FIDO2, MFA (TOTP, recovery codes), OAuth2/OIDC/SAML SSO, session management (token TTL/rotation/revocation), mobile auth hardening (Keychain/Keystore, deep-link safety, attestation), auth threat modeling, auth PR review, step-up auth, and auth bug/token-leak/session fixes. Triggers: "auth", "login", "signup", "passkey", "WebAuthn", "MFA", "2FA", "TOTP", "OAuth", "SSO", "magic link", "password reset", "CSRF", "refresh token", "session fixation", "PKCE", "biometric"

Diagnose and fix CI/CD pipeline failures, test errors, GitHub Actions issues, and code scanning alerts. Use for: red CI builds, GitHub Actions workflow errors, CI test failures (Vitest/Playwright), TS/lint errors blocking builds, failed-check notifications, Supabase/OpenAPI type-gen failures, pre-commit hook failures, GitHub Code Scanning (CodeQL) alerts, and code security vulnerabilities. Triggers: "fix ci", "ci failed", "build broken", "tests failing in ci", "github actions error", "pipeline red", "check failed", "code scanning", "codeql", "security alert", "vulnerability"

Guide for working with the Raamattu Nyt monorepo structure. Use when creating new packages, adding apps, managing workspace dependencies, understanding import patterns, or troubleshooting monorepo issues. Covers npm workspaces, shared packages, cross-app code sharing, and Lovable Cloud deployment.

Expert assistant for analyzing and fixing linting and formatting issues in the KR92 Bible Voice project using Biome and TypeScript. Use when fixing lint errors, resolving TypeScript issues, applying code formatting, or reviewing code quality.

Expert assistant for creating and maintaining admin panel pages in the KR92 Bible Voice project. Use when creating admin pages, building admin components, integrating with admin navigation, or adding admin features. Also use for refactoring admin pages (splitting one page into multiple, extracting components, reorganizing tabs).

Codebase exploration and location finder for the Raamattu Nyt monorepo. Use when finding where specific functionality is implemented, locating constants/tokens/config values, discovering file patterns, or answering "where is X coded?" questions. Helps other skills and agents locate code quickly.

Create reusable React component packages for the Raamattu Nyt monorepo. Use for: extracting a component into a shared package, building a reusable UI library, designing controlled/uncontrolled props contracts, engine abstraction patterns (GSAP, framer-motion), and decoupling components from Supabase/Auth/fetching. Triggers: "create package", "extract component", "make reusable", "shared component", "props contract", "engine abstraction"

Search Engine Optimization and AI Search Optimization (GEO) specialist. Use when: (1) optimizing for traditional search (Google, Bing), (2) optimizing for AI search engines (ChatGPT, Perplexity, Google AI Overviews, Claude), (3) implementing schema markup for AI citation, (4) improving Core Web Vitals (LCP, INP, CLS), (5) creating citation-worthy content structure, (6) zero-click optimization, (7) E-E-A-T signal implementation. Expert in GEO (Generative Engine Optimization), structured data, and AI-readable content.

Expert assistant for designing and implementing subscription features across the Raamattu Nyt monorepo. Use when adding feature limits, implementing quota checks, creating plan-based access controls, building upgrade flows, extending the subscription system, or updating the plans/pricing page (tilaussivu) to reflect which features each plan includes.

Senior principal engineer for Raamattu Nyt codebase. Use for: features with architectural implications (Cinema Reader, search UX, audio), refactoring prototypes to production grade, PR review (architecture/UX/cinema constraints), designing typed APIs that age well, deciding what NOT to build yet, and challenging abstractions / spotting future maintenance risk. Triggers: "code review", "architecture decision", "should this be", "is this portable", "design review", "principal review", "future-proof"

Fetch and process user feature suggestions from the admin panel. Formats suggestions for review, generates AI implementation plans, and implements features after approval. Use for: "/ehdota" (list in_progress), "/ehdota list" (all pending), "/ehdota suunnitelma" (generate plan), "/ehdota toteuta" (implement), "mitä ehdotuksia on jonossa", "käsittele ehdotus", "generoi suunnitelma".

Process IdeaMachina sparks (kipinä) marked for implementation (ai_creatable=true or usage_state='implementing' in ai_prompt.pm_sparks, or feature_suggestions status='in_progress'); for SOIHTU parents, also pulls implementing child sparks. Presents a combined parent+children plan for approval, then routes to the right implementation skill (frontend-design, supabase-migration-writer, edge-function-generator, …). Use when asked to "process sparks", "code kipinä", "toteuta kipinä", "ai-koodaa kipinä", "implement feature suggestion", or similar.

Extract learnings from recent git commits, current conversation, and security reviews. Captures both reactive fixes (what went wrong) and preventive patterns (what to always do). Routes AND fans out each learning to EVERY relevant skill — one canonical entry plus lightweight pointers in each affected skill's learnings.md — so the learning surfaces no matter which skill a future Claude session triggers. Use for: "/opi" or "opi", "what did we learn?", extract/update learnings, end-of-session reflection, after a security review/audit, or after creating reusable components/patterns other skills should know about. Triggers: "opi", "learn", "oppia", "mitä opittiin", "extract learnings", "add learning"

Create and maintain tests for the Raamattu Nyt monorepo using Vitest and React Testing Library. Supports app-specific and package-specific test configurations across the monorepo. Use for: writing tests for hooks/components/utils, analyzing recent commits to find code needing tests, reviewing tests for correctness/coverage, mocking Supabase/auth/other deps, and debugging failing or flaky tests. Triggers: "write tests", "test this", "add tests", "need tests for", "analyze test coverage", "fix failing tests", "mock this", "review tests", "evolution tests", "migration-safe tests", "idea-machina tests"

Idea Machina: Expert assistant for the IdeaMachina app — an idea evolution platform with a 4-stage pipeline: KIPINÄ (ideat) → YDIN (strategia) → SOIHTU (toteutus) → VALO (vaikutus). Features AI-powered spark development, multi-core management, persona generation, activation routing, and progressive stage unlocking. Use for: idea-machina app features (apps/idea-machina/), Evolution system (sparks/cores/force stages; pm_sparks, pm_cores, pm_force_modules, pm_evolutions), AI modules (ai_develop, generate_new_ideas, brainstorm, clarification, activation, persona generation), EvolutionProvider/useEvolution & mutation hooks, legacy ideas (pm_ideas, pm_idea_tags, pm_projects), "Continue to…" workflows, and architecture/data-flow questions. Triggers: idea-machina, evolution, spark, kipinä, ydin, soihtu, valo, core stage, force stage, pm_sparks, pm_cores, pm_evolutions, useEvolution, EvolutionProvider, ideas app, pm_ideas, idea management, continue to project, brainstorm idea, activation mode, kirkasta, core health

Expert assistant for the AI orchestrator system — prompts, features, providers, quotas, usage, and frontend AI UI components. Use when (1) creating or editing AI prompts/templates, (2) configuring AI features and vendor/model bindings, (3) registering new AI features for edge functions, (4) debugging AI quota/403/429 errors, (5) switching AI providers or models, (6) checking AI usage and costs, (7) adding new edge functions that call ai-orchestrator, (8) building AI operation UIs with AIProgressBar, AIProcessingPanel, AIContextPreview, or AIErrorBanner.

Expert assistant for managing AI prompts, features, and configuration in the KR92 Bible Voice AI system. Use when creating AI prompts, configuring AI features, managing prompt versions, setting up AI bindings, or working with AI pricing and models.

Expert assistant for Bible reference lookups and RPC function usage in the KR92 Bible Voice project. Use when fetching Bible verses, generating verse lookups, validating references, working with OSIS format, or choosing optimal lookup patterns.

Expert assistant for keeping documentation synchronized with code changes in the KR92 Bible Voice project. Use when updating API docs, maintaining architecture diagrams, syncing README, updating CLAUDE.MD, or generating documentation from code.