iurFriend

Marketing, web development, and observability skills for Claude Code.

Conversion optimization designer — page audits, page strategy, scorecards, CTAs, trust bars, social proof, and framework-neutral method-lens discipline. Renamed from `conversion-engine` per skills-o6te (2026-05-26).

Brand-scope offer/CTA/journey architecture brain + funnel strategist — offer ladder, CTA cascade, journey mapping, per-funnel projections, and optional clean-room method lenses. Hands off to conversion-designer. Renamed from `funnel-design` per skills-qyv7 (2026-05-26).

Full-lifecycle SEO skill — from keyword research and content planning through editorial optimization to technical audit and post-deploy validation.

Design system generator — visual identities, color palettes, typography scales, accessibility invariants, and modern web platform primitives. Produces DESIGN.md + design-tokens.json.

Rendered page and flow auditor - browser evidence, viewport checks, qualitative UX/CRO/trust findings, impeccable frontend craft checks, privacy-aware screenshots, and remediation routing.

Website and information architecture planner - method selection, top tasks, object/content models, page inventory, form/link-origin maps, funnel-page maps, content hubs, validation plans, Mermaid/Excalidraw sitemaps, and reviewer rubrics. Renamed from `website-architect` per skills-cr9q (2026-07-03).

Concept-to-build architecture plugin - platform/runtime choices, system context, integration maps, ADRs, feasibility gates, and build handoffs.

Management operating plugin - team alignment, ownership maps, cadence plans, decision reviews, communication protocols, learning logs, and motivation/adoption risk.

Management finance plugin - business cases, unit economics, scenario plans, budget gates, runway notes, and investment decision briefs.

Runtime learning-loop plugin - passive measurement analysis, operational-substrate adapter specs, KPI trees, signal reviews, optimisation briefs, governed experiment plans, variant briefs, readouts, and learning logs.

Produces specification.website-aware Astro 6 + Svelte 5 + Tailwind CSS 4 sites from design systems and conversion manifests.

TTS voice assistant with personality — macOS, Gemini, cloud Voxtral, local MLX-Audio (local_tts), and oMLX providers. Optional local (oMLX) summarizer backend for the rewrite hooks, plus per-model TTS lenses (Voxtral/Orpheus/Qwen3/Kokoro). Stop summaries run on a detached announcer so slow local TTS never blocks or times out the hook.

Section-by-section website migration from WordPress to Jamstack. Disciplined extraction-test-implement-verify loop with specification.website-aware migration review. Renamed from `web-refactor` per skills-y43y (2026-06-07).

Langfuse observability operator — CLI-backed API work, traces, sessions, prompts, datasets, scores, tracing guidance, reports, chat extraction, and RAG analysis.

MLflow GenAI observability and evaluation operator — onboarding, tracing instrumentation, trace/session analysis, agent evaluation, metrics, and docs search.

Canonical extraction — produces BRAND.md, VOICE.md, AUDIENCE.md, OFFER.md, COMPANY.md from a brand's source materials, via existing-sources extraction or guided from-scratch elicitation. Bootstraps every other marketing skill once per brand.

Content production meta-bundle. Ships research-analyst, content-strategist (merged with plan), content-writer (merged with draft), editor, deprecate skills plus research-gatherer and content-validator sub-agents. Publishing remains an orchestrator routine plus project-local adapter per R64.

Artifact-level critical-reflection capability per PRINCIPLES P11. Canonical artifact reviewer for greenlit-bound non-draft artifacts, with R52 rule sets for brand-compliance, risk classification, concept-development, solution-architecture, business-economics finance, measurement/optimisation including adapter-boundary review, generated-media with impeccable craft checks, presentation-sibling review, and method-lens misuse/dark-pattern review. Editor remains the specialist looped reviewer for drafts; site-reviewer owns rendered page and flow review. Current plugin/skill: `artifact-reviewer` / `artifact-reviewer`. Renamed from reviewer per skills-3lmx (2026-06-07). Earlier renamed from `quality-assurance` / `qa-assurer` per R54 / skills-rsy9 (2026-05-21).

Front-of-chain concept-development plugin for business concepts. Shapes raw ideas, concept workspaces, management narratives, product or marketplace concepts, campaign/tool launch ideas, and operating-system concepts into concept shape maps, spines, challenge reports, method-lens notes, lane documents, maturity gates, handoff briefs, and learning logs before marketing, PRD, design, or build work begins.

Generative media production — on-brand images, video, product shots, reusable visual style packs, reverse style extraction, trained identities via generative AI (Higgsfield, HeyGen), and animations / motion graphics via a local Remotion studio (deterministic, code-defined motion: kinetic typography, lower-thirds, title cards, explainers, motion data-viz, 3D, Lottie). Orchestration + contract layer: grounds runs in brand/design/conversion artifacts, confirms before spending, and writes durable artifacts to the project media/ zone. The generative sibling of the template-mediated visual-asset bridge.

Rendering presenter — brainstorms, selects, and renders the most impactful visualisations for producer artifacts or ad-hoc data into self-contained (or interactive) reports. Emits a frontmatter-free presentation sibling beside the canonical artifact via presentation-sibling-contract@1 (never indexed in MANIFEST.md, never a status: mutation). A strict presenter: it represents existing values and producer-supplied interpretations and never computes new findings, scores, deltas, rankings, or verdicts. Chart families: line/bar/scatter/pie plus structural no-measure marks — timeline/Gantt (time × category) and presence matrix (category × category).

Portable, composable, multi-channel newsroom. A guided markdown installer stands up a governed, domain-specific newsroom — the wire→dossier→story-arc→commission→edition memory + provenance spine — that produces commissions and campaigns across channels (newsletter, blog, and more). Ports the mechanics, never the content: zero source-installation bleed. The campaign-strategist emits schema-driven stakeholder tables + CSV sidecars and hands off to data-visualization for co-located timeline/Gantt + channel-matrix visuals.

Local podcast studio — turns any source (topic, URL, PDF, markdown, repo) into a finished branded episode: research, multi-voice script, Gemini TTS, music mix, show notes, and cover image. Claude Code is the producer agent running the pipeline locally via Bash; brand-as-config via show profiles, decoupled from any cloud control plane. v0.2 adds a HARD pre-TTS editorial gate, named empathetic format recipes, cast role schema + intro beats, and per-stage QA assertions (incl. a music-present check). v0.3 adds API-economical render paths: R2 duo via Gemini native multi-speaker + R1 solo single-call (one call per topic-chunk instead of per-turn; ~80-90% fewer TTS calls), R3 stays per-turn. v0.3.2 replaces the deterministic env doctor with a markdown-driven first-run onboarding rubric: the agent probes the environment with Bash and installs what it safely can (Python libs, ffmpeg on macOS) with the user's confirmation, while the Gemini key stays a manual out-of-conversation step (R50). v0.4.0 adopts the uv-managed venv convention (R70): the pipeline deps install into a plugin-scoped uv venv (never pip --user/--system into an externally-managed system Python) and every dep-needing pipeline script + check_run.py runs through the resolved venv interpreter; the credential bootstrap (setup_credentials.py) stays on system python3, with a stdlib python3 -m venv fallback for users without uv. v0.4.1 makes pydub survive Python 3.13 (which removed stdlib audioop per PEP 594): the uv venv is pinned to known-good Python 3.12 and the deps carry a conditional "audioop-lts; python_version>='3.13'" backport. Ships the podcast-studio orchestrator skill and a /podcast command.

Active analytics instrumentation plugin - owns the canonical event taxonomy, the tracking plan, and write-side tag/event/custom-dimension/key-event/consent configuration across analytics platforms (GA4, Zaraz, GTM, PageSense), plus live verification that the right events actually fire (pre/post-consent). Ships instrumentation-engineer (plan + apply, write-capable) and instrumentation-auditor (live-firing audit, read-only). Provider-neutral, role-based dispatch over adapters; automation-first (MCP > CLI > API > handoff) behind hard cost/credential/consent gates. The producer that turns measurement-contract@1 requirements into real collected signals for measurement-and-optimisation to read. v0.1.0 - both skills, live GA4 (Admin) + Zaraz (config) adapters, and the live-firing auditor implemented and live-tested; epic skills-jaxw.

Audits and refines recently modified code for maintainability, security, extensibility, readability, and performance/correctness while preserving exact functionality — judgment over a rubric (R32), never a quality score or PASS/FAIL gate. Canonical marketplace adoption of Anthropic's official code-simplifier plugin (Apache-2.0). Use after writing/modifying code, before committing or opening a PR.

Initializes or repairs a project so Codex, Claude Code, and OpenCode share one agent operating layer: Beads task/audit state, AGENTS.md + CLAUDE.md (managed blocks), durable rule docs, and project-local orchestrator/executor/reviewer agents. Deterministic, idempotent bootstrap script; no credentials required. Use to make a repo agent-ready or to repair a drifted operating layer.

Platform-native social-media production and planning. Turns brand canon and upstream content or campaigns into on-brand, per-platform social content — LinkedIn carousels, X threads, Instagram Reel scripts, TikTok hooks, Threads/Facebook posts, YouTube Shorts — and plans the per-channel cadence to ship it. Ships social-strategist (channel strategy, content pillars, a content-calendar-contract@1 calendar that renders through data-visualization, social-context setup, and a social-voice delta brand-dna doesn't hold) and social-producer (platform-native production and atomization, brand-voiced, with media briefs and manual-posting instructions). Consumes brand-dna voice, picks up campaign plans and social-channel commission stubs (aware, not automatic), emits media briefs to creative-media-generation, and feeds measurement-and-optimisation. Credential-free in Phase 1 — produces drafts plus manual-posting instructions; publishing and listening/engagement deferred to later gated phases.

Design fluency for frontend development. 1 skill with 23 commands (/impeccable polish, /impeccable audit, /impeccable critique, etc.) and curated anti-pattern detection.

Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques

Production-grade n8n workflow building skills. Pairs with the official n8n instance-level MCP server to teach Claude how to architect, validate, and ship workflows that work in production.

Skills for the Cloudflare developer platform: Workers, Durable Objects, Agents SDK, MCP servers, Wrangler CLI, and web performance

AI-supervised issue tracker for coding workflows. Manage tasks, discover work, and maintain context with simple CLI commands.

Asana project management integration. Create and manage tasks, search projects, update assignments, track progress, and integrate your development workflow with Asana's work management platform.

Claude Agent SDK Development Plugin

Analyze codebases and recommend tailored Claude Code automations such as hooks, skills, MCP servers, and subagents.

Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.

Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review

Skills for designing and building MCP servers that work seamlessly with Claude — guides you through deployment models (remote HTTP, MCPB, local), tool design patterns, auth, and interactive MCP apps.

Create and edit Obsidian vault files including Markdown, Bases, and Canvas. Use when working with .md, .base, or .canvas files in an Obsidian vault.

Extract clean markdown content from web pages using Defuddle CLI, removing clutter and navigation to save tokens. Use instead of WebFetch when the user provides a URL to read or analyze, for online documentation, articles, blog posts, or any standard web page.

> Use for any codebase exploration or understanding task — reads historical intent from Entire checkpoints instead of guessing. Orchestrates other Entire skills to give the agent provenance-backed answers about code.

Generate accurate, semantic frontmatter for markdown files based on content-type schemas and tag taxonomy. Use when analyzing documents to create frontmatter, enriching vault content, or when user mentions tags/metadata/frontmatter. Specializes in Fabric patterns, blog posts, research notes, client projects.

Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction.

Extract the full design language from any website URL. Produces 8 output files including AI-optimized markdown, visual HTML preview, Tailwind config, React theme, shadcn/ui theme, Figma variables, W3C design tokens, and CSS variables. Also runs WCAG accessibility scoring. Use when user says 'extract design', 'get design system', 'design language', 'design tokens', 'what colors/fonts does this site use', or '/extract-design'.

Use when doing ANY task involving Supabase. Triggers: Supabase products (Database, Auth, Edge Functions, Realtime, Storage, Vectors, Cron, Queues); client libraries and SSR integrations (supabase-js, @supabase/ssr) in Next.js, React, SvelteKit, Astro, Remix; auth issues (login, logout, sessions, JWT, cookies, getSession, getUser, getClaims, RLS); Supabase CLI or MCP server; schema changes, migrations, security audits, Postgres extensions (pg_graphql, pg_cron, pg_vector).