stefan-jansen

This skill should be used when the user asks to "housekeep", "sweep stray transition dirs", "clean up agent-infra debris", "find misplaced .workspace or .claude dirs", or when auditing project trees for the class of debris caused by hooks writing to the wrong project root (nested repos, worker-pool worktrees, stub-only session files, doubled path prefixes). Do NOT use for general project cleanup (see `/system:cleanup`) or for memory garbage collection (see `/memory-gc`).

Write clear, impactful website content for scanners. Use when writing landing pages, service descriptions, CTAs, or above-the-fold content. Covers the 50% test, sentence surgery, power words, page-type guidelines, mobile-first brevity, and techniques for reducing wordiness while maintaining impact.

Match content to what users actually want when they search. Use when creating website pages, blog posts, or any content for search engines. Covers 4 intent types (informational, navigational, transactional, commercial investigation), SERP analysis, query modifiers, and buyer journey mapping.

Technical SEO writing techniques for web content. Use when writing website pages, landing pages, service descriptions, or blog posts for organic search traffic. Covers keyword placement, meta descriptions, heading hierarchy, E-E-A-T principles, internal linking, long-tail keywords, schema markup, and measuring SEO success.

Complete D2 language reference for creating diagrams. Covers shapes, connections, containers, special types (SQL tables, sequences), icons, styling, and advanced features. Use for all D2 diagram generation tasks.

Clarity and simplicity principles for effective diagrams. Focus on essential elements, minimize visual noise, maintain consistency, and use single abstraction levels. Ensures diagrams communicate clearly without clutter.

Visual hierarchy principles for effective diagrams. Use size, color, placement, and typography to guide the viewer's eye and communicate importance. Increases understanding by up to 73%. Apply to all diagrams for clarity.

This skill should be used when the user asks about "CME futures data", "session dates", "Sunday 5pm boundary", "forward returns", or when working with CME futures minute data, handling session vs calendar dates, or filtering invalid timestamps during breaks.

Initialize Claude Code framework infrastructure for any project

Simplification and readability techniques. Use when writing for broad audiences or simplifying complex content. Covers active voice, short sentences, jargon elimination, and accessibility principles from the Plain Language Movement.

Hierarchical content structure - answer first, then supporting arguments, then details

Narrative structure for complex topics - Situation, Complication, Question, Answer

Shared configuration patterns for project setup commands. Provides security hooks, Claude framework structure templates, and framework detection patterns used across multiple setup commands.

API authentication patterns including JWT, OAuth 2.0, API keys, and session-based auth. Covers token generation, validation, refresh strategies, security best practices, and when to use each pattern. Use when implementing API authentication, choosing auth strategy, securing endpoints, or debugging auth issues. Prevents common vulnerabilities like token theft, replay attacks, and insecure storage.

Docker image optimization patterns including multi-stage builds, layer caching, security hardening, and size reduction techniques. Use when building Docker images, optimizing container size, improving build performance, or implementing Docker security best practices. Reduces image sizes by 70-90% and build times by 50-80%.

SQL query optimization patterns including EXPLAIN plan analysis, index strategies, query rewriting, and N+1 query prevention. Use when optimizing slow database queries, analyzing query performance, designing indexes, or debugging database bottlenecks. Works with PostgreSQL, MySQL, SQLite, and other SQL databases. Typical improvements: 10x-1000x query speedup.

This skill should be used at the START of any non-trivial work unit, when the user asks to "align", "spec this out", "define the work", or before planning/implementing. Forcefully interrogates the user (or seeds the spec from a brief/RFC if invoked as `/align @brief.md`) to produce a complete spec.md (WHAT/WHY end-state) that the headless `plan` step can consume without further questions.

This skill should be used when the user asks to "write a handoff", "produce a transition", "record where we are for the next session", "write a digest before /clear", or runs `/handoff`. Writes a durable transition file at `.workspace/transitions/YYYY-MM-DD/HHMMSS.md` that any future session on either host (Claude or Codex) can pick up cold via `/continue`. Host-neutral.

This skill should be used after all issues in a milestone have commits on the milestone's PR branch and the user asks to "ship the milestone", "merge the PR", "close out the milestone", or runs `/ship`. Verifies every milestone issue has a closing-footer commit on the PR branch, marks the PR ready if draft, squash-merges with branch deletion, verifies all milestone issues auto-closed, closes the milestone, and writes a transition entry. Host-neutral — same contract on Claude and Codex.

This skill should be used when the user says "continue from <transition-file>", asks to "resume from the last handoff", "pick up where we left off", or runs `/continue`. Reads a transition file written by `/handoff`, runs its verification snapshot, reports drift, and surfaces the suggested next steps WITHOUT auto-executing. Host-neutral.

This skill should be used after `/plan-issues` has materialized issues on GitHub and the user asks to "implement the next issue", "pick up the next ticket", "do the next one", or runs `/next-issue`. Picks the lowest-numbered open issue in the active milestone, branches if needed, implements per the issue body, runs tests, commits with `Closes

This skill should be used after a `plan.md` exists in an active work unit, when the user asks to "create the issues", "open the GitHub issues from the plan", "materialize issues", or runs `/plan-issues`. Parses `plan.md`, creates the milestone, and opens one GitHub issue per `**Issue N — <title>**` heading. Default is dry-run; pass `--apply` to actually create.