youtrackdb

Review code, test, and workflow-machinery changes across multiple dimensions using specialized agents with triage-based selection.

Research the codebase and create an implementation plan with architecture notes, design document, and track decomposition. Use when starting a new feature or large change.

Execute the implementation plan: autonomous plan review (Phase 2) on first invocation, then track-by-track execution (Phase A review + decomposition, Phase B implementation, Phase C code review). Use after /create-plan to implement the planned work.

Diagnose and fix a CI test failure from a GitHub PR or workflow run. Includes root cause analysis, dimensional code review, and PR creation.

Apply an edit to `design.md` or `design-mechanics.md` through the mutation discipline: apply → auto-review → iterate → present. Use this instead of directly Editing those files.

Audit a finished design document for hard-to-read or hard-to-understand paragraphs, then harden the house-style rules so future design docs avoid them. Fans out audit sub-agents, classifies each obscure passage as caught-by-an-existing-rule or a gap, and proposes plus (on approval) applies rule changes across the style docs. Use when the user wants to feed a design doc back into the style guide, find obscure paragraphs and fix the rules, harden the writing rules from a real DD, or close the readability feedback loop. Accepts a design-doc path, an adr dir name, or defaults to the current branch's design. Reports the doc's own violations; does not rewrite the doc.

Reviews any draft for AI writing tells and produces a clean rewrite. Catches structural fingerprints (negative parallelism like "It's not X, it's Y", bullet-everything, Title Case headings, adjective triads), analysis fingerprints (superficial -ing clauses, passive voice, vague attribution, hedge stacking), and tone fingerprints (throat-clearing, closing phrases like "In conclusion"). Use this skill whenever the user asks to humanize, de-AI, edit, polish, or sanity-check writing, including phrases like "does this sound like ChatGPT", "make this less AI", "humanize this", "edit this draft", "remove the AI smell", "is this AI-y", or "clean this up". Also trigger when the user pastes a draft and asks for feedback on the writing without naming AI specifically.

Manually re-run the autonomous plan review (Phase 2 — consistency + structural). The same review runs automatically as the first phase of /execute-tracks; this command is for re-runs after inline replanning or whenever the plan needs explicit re-validation. TRIGGER when: user asks to re-validate the plan after inline replanning; user requests explicit plan review outside /execute-tracks. SKIP: /execute-tracks already running State 0 in this session (autonomous Phase 2 path will fire) — do not double-run.

Review a workflow-style PR's design, plan, and track files in research-mode Q&A; auto-records observations and submits a line-anchored review via gh api. TRIGGER when: user asks to review a workflow PR or run /review-workflow-pr. SKIP: non-workflow PRs without docs/adr/<dir>/_workflow/.

Migrate a branch's docs/adr/<dir>/_workflow/** artifacts by replaying workflow-format commits from the per-artifact stamp base through HEAD. Resumable across /clear. TRIGGER: branch has stale _workflow/ after a workflow-format change on develop. SKIP: branches with no _workflow/.

Analyze git changes (branch, commit, or current branch) and produce a design-oriented markup document with Mermaid diagrams, query/usage examples, data walkthroughs, PR context, and impact assessment.

Profile and diagnose YouTrackDB SQL/MATCH query bottlenecks. Accepts one or more LDBC query names (e.g., IC5, IS7, IC1,IC10). Combines async-profiler flame graphs, step-by-step selectivity measurement, and fan-out analysis on a Hetzner CCX33 server against the LDBC SF 1 dataset.

Review documentation files for grammar, factual accuracy, and query correctness. Use when the user asks to review docs, validate documentation, or check YQL examples. Accepts a path to a file or directory as argument.

Profile JMH benchmark regressions using async-profiler on a Hetzner CCX33 server. Reads regressions from a PR benchmark comment, profiles both HEAD and BASE with collapsed-stack output, compares self-time and inclusive-time per method, and identifies root causes. Use when the user asks to profile regressions after a benchmark comparison run.

Provision a Hetzner CCX33 server, deploy the project, run JMH benchmarks, collect results, and destroy the server. Use ONLY when the user explicitly asks to run JMH benchmarks on a Hetzner server. Do NOT trigger for general benchmark requests or local benchmark runs.

Generate a PPTX presentation explaining code changes on the current branch, with matplotlib diagrams and dark theme slides.