Web & Frontend

joogy06 Bundle

Use during evo's TESTED phase (modes b/c only) to generate characterization tests by composing integration-flow-testing — never extends or forks it. Three test types — regression-replay (existing pytest/jest baseline), migration-confirmation (api_delta-aware, validates legacy oracle preserved per HARD-RULE 2 bug-for-bug), CVE proof-of-fix (mode-c minimal vuln replay). Every generated test file gets a mandatory header `# CONFIDENCE — characterization-aid | requires-user-review` and pytest.mark.evo_generated / describe.skipIf metadata. Never overwrites existing user tests; only writes to tests/test_evo_<mode>_<component>__<symbol>.py|js paths. Requires a bob-issued claim token, heartbeats every 60s, emits transition requests for TESTED. Skill writes test FILES only; bob's trusted_runner executes them (CB3 provenance).

joogy06 Skill

Use for enterprise Java-shop web front-ends and legacy enterprise single-page apps — Angular (standalone components, signals, RxJS, NgRx state management), OAuth2/OIDC-with-PKCE authentication flows, Nx monorepos, and the browser tier of a Java or Spring enterprise estate. Part of the java-* skill family, companion to java-backend. Trigger on - Angular, NgRx, RxJS, Nx monorepo, OIDC PKCE, enterprise SPA, Java enterprise frontend, JEE or Spring front-end tier.

joogy06 Bundle

Use during bob's Step 4.5 (after UNIT_TESTED) or evo's INTENT_MAPPED phase (per design §6 state machine) to extract per-component functional intent from a project's source code. Anchor-and-expand on contract-map components plus 1-hop call-graph neighbourhoods; per-component LLM pass at temperature=0; structured YAML output conforming to functional-intent.v1 with closed-set function_class enum, REQUIRED evidence_edges that resolve in static.jsonl, two-arm verification on prose claims; content-hash cached. Requires a bob-issued claim token, heartbeats it every 60s, emits transition requests for INTENT_MAPPED. Skill NEVER writes claim files, NEVER writes the integration ledger, NEVER traverses the call graph beyond 1-hop. CB4-compliant.

joogy06 Skill

Use when reading, creating, modifying or securing PDFs from Python or the command line — library and licence selection, text and table extraction from digital versus scanned files, OCR, generating PDFs from code or HTML, merging, splitting, watermarking, encryption and metadata, filling and flattening forms, genuine redaction as opposed to drawing black boxes, PDF/A archival, and treating an inbound PDF as untrusted input. Covers Windows, macOS and Linux including the command-line tools.

joogy06 Bundle

Use when building or debugging modern web front-ends — framework and rendering-mode selection (Vite SPA, Next.js SSR/SSG/ISR, hydration strategies), React/TypeScript component architecture, enterprise SPA patterns (Angular, NgRx, OIDC/PKCE, Nx monorepos), state and server-cache management, responsive layouts, Core Web Vitals as build-time budgets, source-map and browser-devtools debugging, bundle analysis and code splitting, testing (Playwright, MSW, axe/WCAG), and deployment-mode selection (static-first default, SSR only with platform validation). Trigger on - Vite, Next.js, SSR, SSG, hydration, React app, frontend build, web app performance budget, bundle size, frontend debugging, headless frontend. Headless WordPress/WooCommerce data boundaries live in wordpress-developer / woocommerce-developer; CWV measurement lives in performance; trading dashboards live in trading-dashboard-ux; UI/UX design decisions live in audience-experience-design.

joogy06 Skill

Use when building or reviewing React components — Server versus Client Components and where the boundary belongs, Actions and form submission, the use() API, useOptimistic and the other React 19 hooks, the React Compiler and why manual memoisation is now mostly obsolete, state management choices, effects and when NOT to use one, refs, error boundaries, Suspense and data fetching, and the render-behaviour traps that produce stale or duplicated UI.

joogy06 Skill

Use when designing a new component, service, or data flow that crosses a trust boundary, holds secrets, processes untrusted input, or has elevated privileges — produces a structured threat model using STRIDE (per-asset categories), LINDDUN (privacy-sensitive), and kill-chain mapping for high-stakes targets. Outputs a concise, decision-grade threat-model artifact callable by forge Step 1 (security/threat-model branch), alf sweeps, and bob WP-acceptance criteria. Trigger on - threat model, STRIDE, LINDDUN, attack surface, kill chain, attacker model, trust boundary, data flow diagram (DFD), security design review, threat enumeration, "what could go wrong", privacy review, GDPR threat model, MITRE ATT&CK mapping.

joogy06 Skill

Use when building or reviewing a Next.js application on the App Router — routing, layouts and route groups, the rendering modes and where each is appropriate, Cache Components and the opt-in `use cache` model, data fetching and revalidation, Route Handlers and Server Actions, middleware, Turbopack, streaming and Partial Pre-Rendering, environment and secret handling, and the deployment-target decision. Covers the caching inversion that makes most older Next.js advice wrong.

ericrisco Bundle

Use when you have raw 9:16 footage and need it cut into a post-ready vertical short — dead air removed, fast jump cuts, karaoke word-by-word burned-in captions from a transcript, cuts and zooms snapped to the music beat, safe-area placement so captions clear the platform UI, and a platform-correct MP4 export. NOT writing the script, hook, beats or caption text (that is `video-shorts`), NOT building a Remotion React composition codebase (that is `remotion-video`), NOT deciding when or where to post (that is `social-publisher`).

rweisssieker-xp Skill

Use when AI/KI should review Power BI report pages, visuals, axes, KPI cards, slicers, drilldowns, layout density, misleading charts, missing context, accessibility, or decision usefulness.

rweisssieker-xp Skill

Use when a Power BI report should be reviewed like a senior report designer for visual choice, information density, hierarchy, drilldowns, actionability, accessibility, and mobile usability.

planifest Bundle

Next.js App Router implementation workflow for route structure, server/client boundaries, and caching/revalidation behavior. Use when App Router architecture or data/rendering behavior must be implemented or changed; do not use for repository-wide architecture governance or release management policy.

planifest Bundle

Integration-boundary testing for component and service collaboration correctness. Use when modules interact via APIs, queues, databases, or adapters and boundary behavior must be verified; do not use for full UI journey validation or pure unit isolation work.

planifest Bundle

Accessibility-first design workflow for specifying semantics, keyboard/focus behavior, readability, and assistive-technology expectations. Use when UI accessibility behavior must be defined before implementation or design sign-off; do not use for backend data modeling or deployment pipeline decisions.

planifest Bundle

Next.js App Router principles. Server Components, data fetching, routing patterns.

planifest Bundle

React component design workflow for state ownership, composition boundaries, and predictable rendering behavior. Use when React component behavior or structure must be implemented or revised; do not use for repository-wide architecture governance or release policy decisions.

planifest Bundle

C4 architecture modeling workflow for context, container, and component views that make boundaries and responsibilities explicit. Use when teams need a shared structural model before major implementation or refactoring; do not use as a substitute for decision records.

planifest Bundle

Responsive layout design workflow for defining adaptive structure and component behavior across viewport ranges. Use when UI layouts must remain usable and readable across device contexts; do not use for backend data modeling or deployment pipeline decisions.

planifest Bundle

Define scalable design-system foundations with clear ownership and adoption boundaries. Use when multiple teams need shared component standards, foundational patterns, and ownership rules to deliver consistent UI across products; do not use for backend data-model or deployment pipeline decisions.

allinherog-star Bundle

前端审美诊断助手适合内容创作者、市场营销、运营、内容媒体在用户提出“这个界面高级吗”这类问题，需要快速拆解目标、判断重点并形成可执行结果时使用，帮助基于输入材料生成视觉/创意诊断、提示词或分镜方案、发布规格检查。

pooriaarab Skill

Build and submit a Wix app (a React dashboard page under integrations/<name>/ built on the Wix SDK) and list it in the Wix App Market. Use when creating a Wix app, building an embedded dashboard page, wiring Wix OAuth, calling an external API from the dashboard, or fixing a Vite build where import.meta.env is untyped. Covers the whole path plus the traps: a Wix dashboard app is an embedded page (Wix SDK / @wix/dashboard) authorized by Wix OAuth for Wix data — your own API still needs its own key — and a Vite+TS app that reads import.meta.env MUST ship a src/vite-env.d.ts with `/// <reference types="vite/client" />` or tsc errors 'Property env does not exist on ImportMeta'. Sibling of the other integration skills (canva-app, monday-app, shopify-app). Triggers: 'build a Wix app', 'Wix SDK', 'Wix dashboard page', 'Wix App Market', 'import.meta.env has no env property', 'Wix OAuth'.

pooriaarab Skill

Build and submit a Whop app (a React app under integrations/whop-app/ embedded as an iframe view inside a Whop, built on the Whop SDK) and list it on the Whop App Store. Use when creating a Whop app, embedding a tool inside a creator's Whop, wiring 'Sign in with Whop' (OAuth 2.1 + PKCE), calling an external API from the embed, or figuring out the dev-proxy → versioned-build → review submission. Covers the whole path plus the traps: the app runs in a Whop iframe (the Whop user context comes from the SDK, but any OTHER service — your own API — needs its own auth/key, Whop's OAuth does not authorize it), the App ID + redirect are dashboard-owned, and apps can charge money (Whop takes a cut). Sibling of the other integration skills (canva-app, shopify-app, wix-app). Triggers: 'build a Whop app', 'Whop SDK', 'Sign in with Whop', 'publish to the Whop App Store', 'Whop dev proxy', 'my Whop app can't reach my API'.

pooriaarab Skill

Build and submit a monday.com app (a React board/item view under integrations/<name>/ built on the monday apps SDK) and list it in the monday marketplace. Use when creating a monday app, building a board or item view, wiring monday OAuth/session, turning a board row into an external action, or fixing a Vite build where import.meta.env is untyped. Covers the whole path plus the traps: a monday view runs embedded in the board (monday-sdk-js / @mondaycom/apps-sdk gives context + a session token, not authorization for YOUR API), and a Vite+TS app reading import.meta.env needs src/vite-env.d.ts or tsc fails. Sibling of the other integration skills (wix-app, airtable-extension). Triggers: 'build a monday.com app', 'monday apps SDK', 'monday board/item view', 'monday marketplace', 'monday OAuth', 'import.meta.env untyped'.

deciqai Bundle

In a common-value auction where all bidders estimate the SAME uncertain value, the winner is systematically the one who most OVER-estimated it — so winning is bad news that you likely overpaid. The discipline is to shade your bid for the adverse selection of winning, condition your estimate on the fact that you won, and pre-commit to a value-conditioned walk-away ceiling. Activate when: user is bidding in an auction, acquisition, or competitive process for something with a shared uncertain value ('what should we bid', 'we won — did we overpay', 'how high do we go', 'auction', 'best-and-final', 'competitive M&A / acquihire', 'ad auction / max CPC', 'bidding war', 'we keep losing deals — should we bid more'), or someone treats winning a contested bid as pure good news. Do NOT activate when: the value is PRIVATE (tastes differ, so 'you' knowing what it's worth to you is not an estimate of a shared truth) and there is no common-value component; there is a single buyer/seller with no competing bidders; the item's

deciqai Bundle

Activate when: user asks 'how do we prove we're high quality when they can't verify it', 'what credential or signal would work here', 'why is everyone spending so much on brand advertising', 'how do we tell who's the real one among these candidates', 'cheap talk isn't working — we need something credible', or describes a market where one side can't verify the other's quality/intentions/type (hiring, fundraising, B2B procurement, branding, M&A due diligence). Do NOT activate when: information is fully symmetric and both parties already know the relevant facts; or the question is about game theory in general without an information asymmetry component. More: deciqai.com/s/signaling-games

deciqai Bundle

Activate when: user asks 'what will they do if we do X', 'how will competitors react to our pricing', 'how do I design this auction or mechanism', 'we keep ending up in a bad outcome even though everyone prefers better', or is analyzing a strategic situation with multiple rational counterparties (pricing, negotiation, M&A, regulation, platform launch). Do NOT activate when: the decision is essentially solo with no strategic counterparty; the counterparty is clearly irrational or acting on emotion rather than self-interest. More: deciqai.com/s/nash-equilibrium

joogy06 Bundle

Use at forge Step 2.5 for UI designs. Design-phase freeze skill that receives a draft skeleton (from skeleton-extractor), an approved HTML mockup, and user edits; validates every `binds_to: capability://...` URI resolves, enforces D2 strict on unresolved tokens (user must explicitly approve or add each one), HMAC-signs the payload with `.forge/session.key`, and atomically two-file writes `.design-ledger/skeletons/index.yaml` + `<screen>.yaml` via `trusted_runner.bundle_write`. Emits a `skeleton_frozen` transition request (bob applies — CB4). Mirrors S027 test-architect Phase-3 shape.

joogy06 Bundle

Use when you need to show visual content (mockups, wireframes, architecture diagrams, side-by-side layout comparisons) to the user during a design or brainstorming conversation. Generates self-contained HTML files in /tmp/ that the user opens in their browser. Server-less local replacement for the superpowers brainstorming visual companion. Trigger phrases: "show me a mockup", "render this diagram", "visual companion", "compare these layouts visually".

joogy06 Bundle

Merge per-run source artifacts (`static.jsonl` + per-agent `asserted/<agent_id>.jsonl` + optional `manual.jsonl`) into a run-scoped `snapshot.json` conforming to `wiring-snapshot.v1` (v1.1 adds per-component `intent` blocks via `intent_merge`). Single writer of reconciled snapshots. Does NOT promote to `latest.json` — bob does, atomically under flock. Deterministic Python, no LLM calls. Invoked by bob with a claim UUID; heartbeats every 60s via `_meta/claims.py`.

joogy06 Bundle

Use during evo's DRIFT_SURFACED phase (per design §6 state machine) to render diagrams from a frozen intent-map + wiring snapshot + api_delta. Pure-Python deterministic — no LLM calls, no timestamps in payloads, byte-identical output on re-run. Emits 4 diagram types capped at HARD-RULE 5's "3 per consultation turn" — D1 Mermaid sequence (per-component runtime), D2 Cytoscape-JSON blast-radius subgraph, D3 Mermaid sankey for version delta, D4 markdown coverage heatmap. Function-level rendering rejected with explicit error.

joogy06 Bundle

Use when you want a PERSISTENT, queryable, SCIP-inspired code-intelligence library over legacy artifacts — COBOL, DataStage .dsx (XML), ETL scripts (shell/SQL/Python), Java (legacy JEE/Spring estates), and Pick/MultiValue BASIC (UniVerse/UniData/D3/jBASE/OpenQM) — built ONCE into a content-addressed store and exposed via a deterministic graph-query layer + a self-contained HTML navigator. An LLM-as-parser framework (mirrors lineage-extract-static): model-neutral prompts the in-session AI CLI uses to extract symbols/occurrences/relationships, NOT per-format AST parsers. Triggerable as a single skill (ingest one artifact/dir) and via agent-teams (batch, one worker per artifact — NO new agent). Static analysis only (v1). Also trigger on "index this COBOL", "symbol graph for DataStage jobs", "where is this paragraph called", "impact of changing this copybook", "index this Pick/MultiValue BASIC", "index this Java", "code intelligence for legacy", "build a navigator for this mainframe code".

joogy06 Bundle

Use when you want a cheap, standing, read-only GENERATED comprehension front-door (PROJECT.md + per-component COMPONENT.md) for an arbitrary code repository that has no durable contract-map — turns a code tree into non-drifting architecture docs built from a bounded synthetic component partition + structural code edges (AST/regex) + per-component LLM functional intent. Claimless and CB4-safe — writes ONLY to .comprehension/ scratch + generated docs (nothing under .ledger/ or progress/). Composes wiring-extract-static (--standalone), intent-extract (--standalone), the wiring-reconcile pure merge core, and intent-map-render pure functions. Shadow-first - generates to a SHADOW path for side-by-side parity before adoption; --migrate (overwrite real docs) is human-gated and separate. Also trigger on "generate PROJECT.md", "document this codebase", "comprehension front-door", "architecture docs for a repo without a contract-map", "what are the components of this code".

joogy06 Bundle

Use when transforming an HTML mockup into a draft design-skeleton.v1 YAML. Walks the DOM via puppeteer-core at 3 breakpoints, extracts bboxes + computed styles (resolved back to declared tokens) + wired interaction handlers, and emits one draft ready for visual-architect review. Runs the Node subprocess under trusted_runner discipline (CB3-compliant: trusted runner owns execution, skill only generates draft). Invoked by visual-architect at design-phase freeze; NEVER writes the frozen skeleton itself.

joogy06 Bundle

Use when you want to reverse-engineer the SHAPE of data — table column lists+types, COBOL record layouts (PIC/USAGE/COMP, computed byte offsets, REDEFINES, OCCURS), flat-file positional layouts, inferred CREATE TABLE DDL — from legacy artifacts (SQL/DDL, DataStage .dsx, COBOL+copybooks, flat-files). The 3rd lineage-family sibling (flow=lineage-extract-static, symbols=legacy-code-intel, structure=this). A cross-model LLM-as-parser framework: model-neutral prompts the in-session AI CLI runs against chunks via its OWN context — NO per-format parser deps; Python does only chunk I/O, validation, DETERMINISTIC COBOL byte-offset computation, accumulation, rendering. Handles large-file chunking + resumable partial reports; renders HTML/CSV/Excel/wiki/inferred-DDL/OpenLineage SchemaDatasetFacet. Static only (v1). Also trigger on "reverse engineer table structure", "COBOL record layout", "copybook byte offsets", "infer DDL from", "flat-file layout", "schema from DSX".

joogy06 Skill

Use when designing or building the user-facing surfaces of a trading application — order-entry tickets, live blotters, and operator/monitoring consoles — where UX defects convert directly to monetary loss. Covers order-ticket safety (side/size/symbol/notional review, exact-order-digest confirmation, fat-finger guards, keyboard and focus safety), order-lifecycle presentation without optimistic success, market-data freshness encoding (quote age, feed degradation, session/halt/connection state), blotter and partial-fill presentation, alert prioritization, and render contracts that keep emergency cancel/reduce/flatten controls responsive under tick storms (off-main-thread ingestion). Trigger on - trading dashboard, order ticket UI, order entry form, live blotter, trading terminal UX, stale quote display, emergency flatten button. Analytical charts and KPI tiles live in the dataviz skill; generic SPA mechanics live in modern-frontend; signal design lives in observability.

joogy06 Skill

Use when slide outlines, visuals, and styling are ready to compile into final PPTX and/or HTML output — export, "generate the file", "give me the PPTX", create HTML slides. Part of the presentation-* skill family.