slogsdon

Exports document-flow HTML to a PAGINATED PDF at a real paper size (Letter/A4), with a running folio footer (title · page X / Y). The long-form sibling to pdf-html: where pdf-html renders each .canvas to one tall auto-height page (right for social artifacts and audit reports), this paginates a whole document across pages — right for essays, field guides, specs, and anything meant to be read or printed as pages. Trigger on: "paginated PDF", "print PDF", "PDF with pages", "Letter/A4 PDF", "multi-page PDF", "PDF of this essay/guide/doc". For single-canvas design artifacts (LinkedIn, Instagram, business card, audit report), use pdf-html or screenshot-html instead.

Generate a blog post hero artifact — TWO pixel-exact HTML canvases in a single file: an Open Graph card (1200×630) for social meta tags AND an in-page hero (1440×600) for the article header. Both canvases share the same content + brand tokens; only the proportions differ. Reads ./design/<brand-slug>/DESIGN.md, tokens.css, and components.html. Triggers include "blog hero for X", "OG card for X", "open graph image", "in-page hero", "/blog-hero".

Generate a printable business card — pixel-exact 1050×600 front canvas + 1050×600 back canvas in a single HTML file. Standard 3.5×2 inch card at 300dpi. Front carries name + title + contact details; back is wordmark-dominant (or brand device) with minimal text. The design system's most reduced artifact — tests whether the brand can speak with almost no text. Print bleed instructions in footer comment. Reads ./design/<brand-slug>/DESIGN.md, tokens.css, and components.html. Triggers include "business card", "calling card", "name card", "/business-card".

Generate a LinkedIn / Instagram multi-slide carousel — single HTML file with N square 1080×1080 .canvas divs, each representing one slide. Default N=5 (cover + 3 content + closing CTA). Cover and closing always use the brand's most distinctive treatments; middle slides are content-led. Reads ./design/<brand-slug>/DESIGN.md, tokens.css, and components.html. Triggers include "carousel for X", "linkedin carousel", "instagram carousel", "carousel slides", "/carousel-slide".

Generate an Instagram post artifact — pixel-exact HTML canvas at 1080×1350 (portrait, default — Instagram's max-real-estate format) with opt-in 1080×1080 (square) and 1080×1920 (story / reel cover) modes. Reads ./design/<brand-slug>/DESIGN.md, tokens.css, and components.html. Triggers include "instagram post about X", "ig post", "ig story", "instagram artifact", "/instagram-post".

Generate a deliverable landing-page HTML file (NOT a screenshot artifact). 1440px design width, mobile-responsive (CSS-only, no JS). Sections: hero (full-viewport) + 3 feature blocks + CTA. Designed to be opened in a browser, iterated on, and deployed as-is. Reads ./design/<brand-slug>/DESIGN.md, tokens.css, and components.html. Triggers include "landing page for X", "marketing page", "product landing", "/landing-page".

Generate a minimal single-column standalone HTML page — mobile-first at 375px width — for use as a link-in-bio destination. Contains: name/wordmark, short tagline, list of links (each a styled <a> block), social footer. Self-contained, works offline, deployable as-is to GitHub Pages or Netlify. Reads ./design/<brand-slug>/DESIGN.md, tokens.css, and components.html. Triggers include "link in bio", "linktree replacement", "bio page", "/link-in-bio".

Generate a LinkedIn post artifact — a 1200×627 image canvas (LinkedIn's recommended share image dimension) plus a companion text suggestion (1300-char limit, hook-first structure). Reads ./design/<brand-slug>/DESIGN.md, tokens.css, and components.html. Triggers include "linkedin post about X", "linkedin artifact", "generate a linkedin image", "/linkedin-post".

General-purpose Obsidian vault skill — use this whenever the user wants to do anything with their vault or personal notes. Covers reading notes by name, searching vault content, creating new notes, appending to existing ones, checking backlinks, and working with daily notes. Trigger immediately on phrases like: "look up my note on X", "add this to my vault", "what does my note say about Y", "find notes about Z", "log this to my vault", "update my daily note", "create a note for...", "what links to X", "search my vault for...", "show me what I know about...", "check my notes on...", or any time the user mentions Obsidian, their vault, or personal notes. If the user is asking about something they might have written down, reach for this skill first.

Use when asked to audit backlinks or repair the knowledge graph structure, or when /backlinks is invoked. Do NOT use for finding conceptual bridges between ideas — use /connect for that.

Map the branching question space from a topic — sub-questions, adjacent domains, meta-framings. Use when /bloom is invoked or Shane asks "what questions does X open up", "bloom on X". Do NOT use to trace how thinking evolved — that's /trace.

Use when Shane wants to pressure-test a belief, steelman an opposing view, or invoke /challenge. Do NOT use for confirming or validating a position.

Use when asked how knowledge has accumulated or compounded around a topic, or when /compound is invoked. Do NOT use for tracing evolution of thinking over time — use /trace for that.

Surface non-obvious cross-references between vault notes — connecting two concepts that aren't already explicitly linked. Use when /connect is invoked, when the user says "what does X have to do with Y", "is there a thread between these notes", "what connects these ideas", or when reviewing two superficially unrelated topics for shared structure. Do NOT use for direct backlinks (use /backlinks) or for surfacing patterns within a single topic (use /emerge).

Use when asked to find contradictions or inconsistencies in thinking, or when /contradict is invoked. Do NOT use for pressure-testing a single belief — use /challenge for that.

Analyze gaps between stated priorities (OKRs, focus statements) and actual session behavior to surface drift. Use when /drift is invoked, when the user says "am I doing what I said I would", "where am I off-track", "do my actions match my goals", "am I avoiding something", or during weekly retrospectives. Reads accountability.md, patterns.md, and recent daily notes. Do NOT use for surfacing patterns within a topic — use /emerge for that.

Find keyword cannibalization in Google Search Console data — multiple URLs from the same site competing for the same queries. Use when the user says "find cannibalization", "are my pages competing", "which pages overlap", "keyword cannibalization audit", "why do my rankings keep flip-flopping", or pastes/uploads a GSC query+page export and asks which URLs fight each other. Works standalone from a GSC CSV; no external API required.

Detect content decay — pages that have lost organic traffic versus a past peak. Use when the user says "find decaying pages", "what traffic did I lose", "content decay audit", "why is traffic dropping", "which posts are declining", or pastes two GSC exports (current vs. an earlier/peak period) and asks what slipped. Works standalone from two GSC CSVs.

Audit whether the queries a page ranks for actually match what the page is about. Use when the user says "intent mismatch", "audit search intent", "why is this page not converting", "are my pages matching what people search", "wrong queries ranking", or pastes GSC query+page data plus a page title/URL and asks if they line up. Works standalone from a GSC CSV plus page titles the user provides.

Find internal-linking opportunities to push authority from a strong "donor" page to a weaker "target" page. Use when the user says "internal links to boost a page", "reallocate link equity", "which pages should link to X", "help page B rank by linking from A", "internal link audit for a target page", or names a donor and target page with GSC data. Works standalone from GSC exports the user provides.

Cluster a large GSC query list into topics and search intents, then map each cluster to a page or flag it as a content gap. Use when the user says "cluster my keywords", "group these queries", "topic clusters from GSC", "organize my search queries", "what content gaps do I have", "keyword grouping", or pastes a big query export. Works standalone from a GSC CSV.

Detect sudden ranking or traffic anomalies in GSC time-series data — big position swings or click spikes/drops worth investigating now. Use when the user says "any ranking anomalies", "did anything spike or crash", "alert me to big changes", "what moved suddenly", "ranking volatility check", or pastes GSC data with a date dimension covering 30+ days. Uses DataForSEO to correlate with SERP/algorithm context when available; works standalone from a GSC CSV.

Map a competitor's ranking content for a topic and find the gaps you can exploit. Use when the user says "spy on a competitor", "what content does competitor X rank for", "competitor content gap analysis", "what's competitor.com ranking for in topic Y", "map their SERP coverage", or names a competitor domain plus a topic. Uses DataForSEO when available for live SERP/ranked-keyword data; otherwise works from SERP results the user pastes.

Find "striking distance" keywords — queries ranking on page 2 (positions 11–20) that are one push away from page-1 traffic. Use when the user says "striking distance keywords", "page 2 keywords", "quick SEO wins", "low-hanging fruit", "what's almost ranking", "where's my fastest traffic upside", or pastes a GSC query export and asks what to optimize next. Works standalone from a GSC CSV.

Use when /eod-pro is invoked, or when Shane says "professional eod", "work eod", "close out my work day", "end of work day", or wants to run his end-of-day accountability audit against the PROFESSIONAL (Global Payments / WorldPay) vault and its GP OKRs. Diffs Today's Focus vs Session Log, increments deferrals in the Professional patterns.md, flags 3+ deferrals, and writes an EOD Audit to today's Professional daily note. For the Personal vault use /eod instead.

Use when /log-pro is invoked, or when Shane says "professional log", "work log", "log this to work", "log to professional", or wants to capture a lightweight session note in the PROFESSIONAL (Global Payments / WorldPay) vault. Appends a timestamped entry to today's Professional daily note under Session Log. For the Personal vault use /log instead.

Use when /morning-pro is invoked, or when Shane says "professional morning", "work morning", "start professional day", "start my work day", or wants to start his GP/WorldPay work day with a focus review against the PROFESSIONAL vault's GP OKRs. Reads Professional accountability context and the most recent Professional daily note, surfaces carryovers, and writes Today's Focus to today's Professional daily note. For the Personal vault use /morning instead.

Use when Shane is stuck on a decision, wants to cut through avoidance, or invokes /decide. Takes a stuck decision as input, loads accountability + patterns context, and outputs a structured session — Mirror (what you're actually doing) → Real Decision → Recommendation → Next Step. Hard mode strips diplomatic framing. Trigger on /decide, "I keep going back and forth on", "I can't decide", "help me think through", "I've been avoiding", or any stuck decision framing.

Use when /eod is invoked or when Shane wants to close out his day. Diffs Today's Focus vs Session Log, increments deferrals in patterns.md for unlogged tasks, flags items at 3+ deferrals, and writes an EOD Audit to today's daily note.

Use when /plan-tomorrow is invoked or after /eod. Reads today's EOD audit and proposes tomorrow's focus (1 primary, 2 secondary) accounting for known patterns. Writes a Tomorrow block to today's daily note.

Use when /morning is invoked or when Shane wants to start his day with a focus review. Reads accountability context and yesterday's daily note, surfaces carryovers, runs a knowledge briefing, and writes Today's Focus to today's daily note.

Process the Obsidian Inbox interactively — reads all files, generates routing recommendations, then walks through each item one at a time with user confirmation before acting. Use when /inbox-process is invoked or when Shane wants to clear the Inbox.

Orchestrate external agent CLI harnesses (pi, hermes) in turn-by-turn sessions from within Claude Code. Use when the user wants to delegate tasks to pi or hermes, chain multiple agent turns, compare outputs across harnesses, or build multi-agent workflows where Claude acts as the coordinator.

End-to-end publishing pipeline for a new article on shane.logsdon.io. Runs text quality passes (humanize → ms-style-pass), generates brand-system design artifacts (blog hero OG + in-page hero, LinkedIn post image), screenshots them at 1x, wires images into the article frontmatter and site, saves LinkedIn companion text to Obsidian, builds the site, bumps the service worker cache version, and commits + pushes. Unattended — no human checkpoints. Working directory must be the site repo root. Triggers: "publish <slug>", "publish the post", "run the publish pipeline", "/publish-post".

Run the monthly Claude Code skill hygiene pipeline — skill-audit (rank skills by usage + quality) → consolidate-memory (prune stale memories) → vault-lint (catch orphans created by deprecated workflows). Use when the 1st of the month rolls around, when the user says "monthly skill audit", "is my context layer healthy", "skill hygiene check", or after deprecating a workflow / removing a skill. Do NOT use for single-skill review (read the SKILL.md directly) or in the same week as the last audit. Output goes to ~/Downloads + Inbox/.

Run the full weekly vault-health pipeline — vault-lint (consistency check) → backlinks (graph repair) → vault-index (rebuild overview) → weekly-signals (deferral / pattern surface) → weekly-learnings (publishable thread surface). Use when /weekly-review is invoked, on Friday afternoons, when the user says "weekly vault review", "audit my vault", "what should I publish this week", or after a week of heavy capture activity. Skips stages whose preconditions aren't met (e.g. weekly-learnings needs ≥3 daily notes from the week).

Draft a blog post or article and prepare it for publishing — idea → first draft → humanize (remove AI-typical patterns) → ms-style-pass (term + bias + heading conventions). Use when the user says "draft a post about X", "write up this idea as a blog", "prepare a draft on Y", "stage a draft for publishing", or after /learned surfaces a publishable thread. Hands off to publish-post for the heavyweight pipeline (artifact generation + publishing); does NOT publish itself.

Run the full feature-spec pipeline for a new product / feature with commercial stakes — biz-brief (business context) → superpowers:brainstorming (technical spec) → superpowers:writing-plans (implementation plan). Use when the user says "let's build X", "I want to ship Y", "spec out this feature", "go from idea to plan for Z", or when starting any new feature with a user / market / business model. Do NOT use for purely internal tooling (skip biz-brief; go straight to brainstorming) or for bug fixes (use debugging-and-error-recovery instead).

Run the full design pipeline for a brand or one-off artifact — design-plan (strategy) → design-system (tokens + showcase) → an artifact skill (linkedin / blog-hero / etc.) → screenshot-html (PNG export). Use when the user says "design a brand", "I need the full design system + assets", "set up a new brand from scratch", "design pipeline for X", or when starting visual work on a brand that has no DESIGN.md yet. Do NOT use when the brand already exists and you just need one artifact — call the artifact skill directly.

Generate or update an AGENTS.md / CLAUDE.md / context file for a repo. Use when the user says "set up CLAUDE.md", "generate AGENTS.md", "write a context file for this repo", "document this codebase for AI agents", or when a fresh repo has none. Also use when an existing context file has gone stale and needs a rewrite from current source. Do NOT use for editing project-specific README or contributor docs.

Secure authentication and authorization patterns — password hashing, brute-force protection, session vs JWT, OAuth/SSO flows, RBAC. Use when the user says "add login", "set up auth", "is this auth secure", "implement OAuth", "session vs token", "should I use JWT here", or when shipping any feature that gates access. Do NOT use for guessing at credentials or bypassing auth — refuse those.

Modern CSS architecture and design systems without heavy frameworks — Grid, Flexbox, custom properties, container queries, layer organization. Use when the user says "set up the CSS for this", "design system", "CSS architecture", "no Tailwind", "vanilla CSS only", or when starting a stylesheet from scratch on a new project. Do NOT use for utility-class systems (Tailwind etc.) — use those frameworks' own conventions.

Relational database schema design, normalization, indexing, and query optimization. Use when the user says "design the schema for X", "how should this table be structured", "this query is slow", "do I need an index here", "denormalize this", or when starting a new feature that needs persistence. Covers Postgres, MySQL, SQLite. Do NOT use for ORM-specific syntax — covers conceptual design first, ORM second.

PHP / Composer dependency management and security auditing. Use when the user says "audit this composer.json", "is this package safe", "update dependencies", "find vulns", "lock this version", or before merging anything that adds a new package. Do NOT use for npm / pip / cargo dependencies — this is Composer-specific.

PHP 8.2+ project organization and architecture with PSR-4 — directory layout for MVC, DDD, microservices. Use when the user says "structure this PHP project", "organize my PHP code", "set up PSR-4", "where should this class go", or when starting a new PHP repo. Do NOT use for framework-specific scaffolding (Laravel, Symfony) — use those frameworks' own generators.

TypeScript configuration, type-safe patterns, and migration from JavaScript. Use when the user says "set up TypeScript", "migrate this JS to TS", "tighten my tsconfig", "fix these type errors", "design types for X", or when a project introduces TS for the first time. Includes generics, conditional types, and JSDoc-typed JS workflows. Do NOT use for general JavaScript questions — this is for type-system work specifically.

Modern vanilla-JS architecture without framework dependencies — ES6 modules, custom elements, state without React/Vue. Use when the user says "no framework", "plain JS", "vanilla JavaScript", "I don't want React for this", or when starting a small static site / utility / progressive-enhancement layer. Do NOT use when a framework is already chosen — use the framework's own conventions instead.

Web standards compliance and progressive enhancement — semantic HTML, ARIA, viewport / form best practices. Use when the user says "make this accessible", "is this semantic", "fix the a11y", "progressive enhancement", "noscript fallback", or when reviewing markup before ship. Do NOT use for visual design — pair with a css-architecture or design-* skill.

Use when about to write a long prompt, include verbose text in context, or when any block of text needs to be made token-efficient before use. Delegates compression to a local model via the LiteLLM proxy (ollama-agent MCP).

Use this skill to delegate execution-level tasks to a local model via the LiteLLM proxy (default alias `fast-general`, resolving to gemma4:e4b-mlx). Best for: vault skill invocations (/ghost, /challenge, /emerge, /contradict, /drift, /ideas, /trace, /connect, /compound, /bloom, /stranger, /map, /level-up, /learned, /weekly-learnings, /backlinks), file search and summarization, drafting content, and any task where local execution is sufficient and API cost should be minimized. Do NOT use for tasks requiring real-time web access, complex multi-step tool use, or high-stakes decisions.

Guide writing a new skill file following Shane's authoring conventions — principle-first, anti-patterns before steps, _lib/ references over inline duplication. Use when invoked as /write-a-skill or when Shane says "create a new skill for X", "add a skill", "write a skill that does X".

Edit a draft to sound like Shane wrote it — removing AI-typical patterns, restoring conversational flow, and preserving technical precision. Use whenever Shane says: "humanize this", "edit in my voice", "AI detector pass", "make this sound like me", "rewrite this", "edit this post", "clean this up". Also invoke proactively if presented with a draft that contains staccato fragments, dramatic colon-reveals, em-dash lists, or semicolons between independent clauses — these are reliable markers that text needs a voice pass.

Apply Microsoft Writing Style Guide term preferences, bias-free language rules, and heading conventions to a blog draft — without disrupting voice or structure that humanize already established. Use after the humanize skill when Shane says: "ms style pass", "apply the style guide", "style check this", "run the style guide on this", "check for outdated terms", "bias check this", "clean up the terminology", "add headings". Also invoke proactively when a humanized draft contains terms like blacklist, whitelist, master/slave, sanity check, guys, or disability metaphors.