svetzal

Apply consistent, humane UX principles when designing, building, or reviewing command-line tools — command grammar and subcommands, verb and flag naming, plan/apply safety, human vs --json output, help text, error messages, install footprint, and client/daemon splits. Use this skill whenever the user is creating a new CLI tool or adding commands, flags, or output to an existing one; choosing verbs or flag names; designing a tool's --help, dry-run, confirmation, or error messages; or doing a UX/design review of a CLI's surface — even if they don't say "UX". Trigger on mentions of CLI design, subcommand structure, flag conventions, argparse/click/clap/cobra/oclif command surfaces, dry-run vs apply, --json output, shell completions, exit codes, or "review my CLI".

Convert a bitmap/raster image (PNG, JPEG, GIF, BMP, TIFF, WebP) into a clean vector file (SVG, PDF, EPS, DXF, or PNG) using the vectorizer.ai API. Use this skill whenever the user wants to vectorize an image, trace a logo, turn a PNG/JPG into an SVG, get a scalable version of a raster graphic, clean up a pixelated logo, prep artwork for print or large-format display, or asks to "convert this to SVG/vector". Trigger even if they don't say "vectorize" — e.g. "make this logo scalable", "I need an SVG of this icon", "trace this bitmap", or "turn this screenshot of a logo into something crisp". Spends vectorizer.ai credits (1 per production render), so it also covers free test renders and cost-aware previews.

Stacey Vetzal's blog writing voice and style guide. Use this skill whenever writing blog posts, articles, or long-form content for Stacey Vetzal, stacey.vetzal.com, or "Stacey on Software". Also use when the user asks to write in Stacey's voice, draft a post, ghostwrite for Stacey, or when any content needs to sound like Stacey wrote it. If you're writing prose that will be published under Stacey's name, this skill applies.

Generate cyberpunk-styled images for blog posts including banners (16:9), callouts (1:1), and diagrams (9:16). Use this skill when creating new blog posts, when asked to generate images, or when updating visuals for existing posts. Creates consistent character-based imagery following the blog's visual identity.

Apply human-centered visual design principles when building, editing, reviewing, or auditing any UI — web pages, dashboards, components, forms, emails, documents, slide decks, or markdown rendered to HTML. Use this skill whenever you're producing or critiquing something a human will look at, especially when generating HTML/CSS/Tailwind, React/Vue/Svelte components, Jinja or ERB templates, design tokens, or styled output. Also use when the user asks things like "make this look better," "clean up the layout," "review this design," "this feels off," "polish the UI," or describes a screen feeling cluttered, busy, untrustworthy, amateur, or hard to scan. Trigger even if the user doesn't say "design" — building UI is design. The skill encodes the spacing, hierarchy, typography, color, accessibility, and motion invariants humans subconsciously use to judge whether a screen feels coherent and trustworthy.

Generate images for Slidev presentations including covers (16:9), section breaks (16:9), concepts (1:1), backgrounds (16:9), and accents (1:1). Abstract-first approach — character is OFF by default. Supports multiple visual styles via scene variants. Use when creating presentation imagery, generating slide visuals, or adding custom graphics to a talk.

Python dependency management with pip and venv. Use this skill when working in a Python project that uses pip, requirements.txt, setup.py, or pyproject.toml without uv.lock. Also use when the user mentions pip, venv, virtualenv, or asks about Python dependency management in a traditional pip-managed project. Provides dependency management patterns and quality gate commands for pip workflows.

Python project management with uv — the fast, Rust-based package and project manager. Use this skill when working in a Python project that has uv.lock, .python-version, or when pyproject.toml uses [tool.uv] or [dependency-groups]. Also use when the user mentions uv, "uv run", "uv add", or asks about Python dependency management in a uv-managed project. Provides project setup, dependency management, running tools, and quality gate commands specific to uv workflows.

Drive cmx, the package manager for curated agentic context (agents and skills) across AI coding assistants. Use this skill whenever the user mentions cmx, context mixer, installing or updating agent/skill files for Claude Code/Copilot/Cursor/Windsurf/Gemini CLI/opencode/Codex CLI/Pi/Crush/ Amp/Zed/OpenHands/Hermes/Devin, cmx sources, cmx doctor, adopting orphaned skills, reconciling drifted skills across tools, or managing a canonical home for hand-authored agents/skills. Also use when the user asks "what agents/skills do I have installed", "is anything out of date", "add this repo as a source", or wants to survey/clean up a messy multi-tool AI assistant setup. Even if the user doesn't say "cmx" by name, use this skill whenever the task is installing, updating, listing, searching, or auditing agent/skill files that AI coding assistants read from disk.

Fixture skill

Fixture skill

Fixture skill

Fixture skill

Fixture skill

The dealer for a repo-grounded Elevation of Privilege (EoP) threat-modeling game night. Use this skill when the user wants to "deal an EoP deck for <repo>", run an EoP / STRIDE game session against a real codebase, produce a printable or web threat-modeling deck where each card names real files, endpoints, and trust boundaries, or get a dealer's diagram plus a score card to play by the original rules. Trigger on "deal me an EoP deck", "run a threat-modeling game on this repo", "make a STRIDE card deck for our system", or "set up an EoP session". Builds on eop-deck.

Shared foundation for the Elevation of Privilege (EoP) threat-modeling skill suite. Use this skill whenever you need the EoP / STRIDE deck data, need to ground a repo or design against the 74 STRIDE threat cards, need to draw a trust-boundary / data-flow diagram for threat modeling, or need to apply the "would we take a bug for that?" adjudication test. Also use when another EoP skill (eop-deal, eop-referee, eop-sentinel, eop-race, eop-forge) needs the deck or a grounded deck object. Trigger on mentions of STRIDE, Elevation of Privilege, threat cards, trust boundaries, data-flow diagrams for security, or "ground these cards against my code."

The design-time Elevation of Privilege (EoP) skill — threat-model a system before the code exists, and extend the deck with an AI/agent suit. Use this skill to "threat model this design", walk STRIDE against a proposed design, RFC, or architecture sketch and write the threats into the design doc before implementation (shift-left), or to "add the AI suit" — a co-authored seventh suit covering prompt injection, tool/agent abuse, training-data and context poisoning, and over-broad agent permissions (a literal elevation of privilege). Trigger on "threat model this design", "shift-left threat modeling", "model this RFC for security", "walk STRIDE against our design", "add the AI/agent suit", or "threat model our agent/LLM system". Builds on eop-deck; emits an eop-session/1 that eop-referee can turn into backlog.

The adversarial Elevation of Privilege (EoP) skill — race the machine to the novel attack. Use this skill to run a timed contest where the table and an adversarial agent each generate novel exploit hypotheses (Ace-class — the kind with no CVE yet) against a design or repo, blind to each other, then reveal and score: the table scores for threats it names before the agent surfaces them, the agent scores for what it finds first, and the visible speed gap is the lesson. Trigger on "race my design", "race this repo", "beat the machine to the threat", "run an EoP race", or "generate novel attacks against this design on a timer". Builds on eop-deck; hands accepted attacks to eop-referee.

The adjudicator and riff engine for a live Elevation of Privilege (EoP) threat-modeling session. Use this skill during play to rule on a contested threat with the bug-test ("rule on this threat: …"), to unstick a player frozen on an Ace by seeding three novel-attack starting points, to run the 60-second riff variant, or to convert accepted threats into backlog — ticket titles, repro sketches, candidate fixes and tests. Trigger on "rule on this threat", "is this a real threat?", "we're stuck on the Ace", "riff on this", "turn these threats into tickets", or "referee our EoP game". Builds on eop-deck and pairs with eop-deal.

The Elevation of Privilege (EoP) game that never ends — embedded in the pull request. Use this skill to "play a hand" against a diff or PR: pick the highest STRIDE card the change legitimately invites, comment in the card's voice (threat, why-here, candidate fix), and track clean-diff streaks plus a "hardest to threaten" leaderboard so the threat-modeling vocabulary stays alive between sessions. Trigger on "play a hand against this PR", "run the EoP sentinel on this diff", "STRIDE-review these changes", "what threat does this change invite?", or "update the sentinel leaderboard". Builds on eop-deck; complements eop-deal and eop-referee.

Local-only, privacy-first personal finance CLI. All mutations dry-run by default; pass --write to persist. Use for import, categorization, budgeting, duplicates, and reporting.

Local-only, privacy-first personal finance CLI. All mutations dry-run by default; pass --write to persist. Use for import, categorization, budgeting, duplicates, and reporting.

Use Epilogue Tracker (et) for user-centric product planning with the Screenplay Pattern. Use this skill whenever the user mentions actors, goals, interactions, journeys, the Screenplay Pattern, product planning, work breakdown, who benefits from work, product charter, or any et CLI commands. Also use when the user asks about structuring work around user needs rather than tasks.

Run iterative codebase improvement with hone CLI; use when user asks to hone, iterate, maintain dependencies, assess, improve code quality, set up or fix quality gates, derive agents, mix agents, or check iteration history

Create cross-agent compatible skills with SKILL.md format, supporting files, and scripts for Claude Code, Copilot, Cursor, Windsurf, and other AI coding assistants

Set up and manage researcher document repositories; use when adding repos, indexing documents, checking status, configuring embedding providers, or running the MCP server

Search indexed document repositories with semantic search; use when looking up notes, documentation, or knowledge base content, or finding relevant context during coding

Set up or troubleshoot GitHub Actions workflows for releasing cross-platform Qt applications. Use when building, signing, notarizing, or packaging Qt apps for macOS, Linux, or Windows.

How to use the Foundry workflow engine for engineering automation. Use this skill whenever the user mentions foundry, foundryd, foundry iterate, foundry scout, foundry validate, foundry run, foundry pipeline, foundry release, foundry gates, foundry campaign, campaigns, foundry registry, foundry sentinel, sentinels, scheduled triggers, quality gates, maintenance runs, drift assessment, pipeline health, CI remediation, release automation, or wants to automate code quality workflows across projects. Also use when the user asks about .foundry directories, trace files, audit reports, event-driven workflows, managing a registry of software projects, or checking what happened in a previous foundry run. Even if the user doesn't say "foundry" explicitly, use this skill when they ask about automated iterate/maintain cycles, gate health checks, or agent-driven releases.

Dispatch concrete, ready-to-execute work to background Hopper workers via the hopper queue. Use this skill when the user wants to queue up substantive coding tasks for unattended processing in specific projects on their machine — not for planning, to-do tracking, or lightweight tasks.

Email management via the mailctl CLI — search, read, organize, and manage receipts across IMAP accounts

Use Alloy (the `alloy` CLI) to capture and refine engineering intent as six-field records — capability, threat, expectation, strategy, evidence, tradeoff. Use this skill whenever the user mentions engineering intent, intent records, capability/threat/expectation/strategy/evidence/tradeoff, why a design decision was made, preserving an architectural capability, a product charter, the Foundry execution engine, or any `alloy` CLI commands. Also use when the user wants to record the reasoning behind an engineering judgement so agents can preserve it.

Run iterative codebase improvement with hone CLI; use when user asks to hone, iterate, maintain dependencies, assess, improve code quality, set up or fix quality gates, derive agents, mix agents, or check iteration history