Claude Skills by darkroomengineering

Feature implementation workflow. Use when the user wants to: - "build a", "create a", "implement", "add feature" - build a new page, component, feature, integration - create something from scratch - add new functionality to the app

Create React components following Darkroom standards. Use when: - User says "create component", "new component", "add component" - User wants to build a UI element, widget, or reusable piece - User mentions a component name like "Button", "Header", "Card"

Codebase exploration and understanding. Use when the user asks: - "how does X work?", "where is X?", "find X", "understand X" - "what files handle X?", "show me the X implementation" - "navigate to X", "explore the codebase" - any question about code structure, architecture, or implementation details

Bug investigation and resolution workflow. Use when the user mentions: - "fix", "broken", "not working", "bug", "error", "issue", "failing" - debugging, troubleshooting, investigating problems - something that used to work but doesn't anymore - errors in console, build failures, test failures

Create custom React hooks following Darkroom standards. Use when: - User says "create hook", "new hook", "custom hook" - User wants to extract logic into a reusable hook - User mentions "use" prefix like "useAuth", "useScroll"

Multi-agent task orchestration for complex features. Use when: - User has a complex, multi-step task - Task requires multiple agents working together - User says "coordinate", "orchestrate", "complex task" - Work involves planning, implementation, testing, AND review

Visual QA validation with agent-browser. Use when: - User asks to "validate", "check visually", "QA check" - After creating or modifying components - Before PR submission - User mentions "accessibility", "a11y", "contrast", "touch targets" - User wants to verify UI looks correct

Code refactoring workflow. Use when: - User says "refactor", "clean up", "reorganize", "restructure" - Code needs improvement without changing behavior - User wants to improve code quality, readability, or performance - Technical debt needs addressing

Code review for quality, security, and best practices. Use when: - User asks to "review", "check", "look at" code - Before merging, committing, or shipping - User mentions PR, pull request, changes, diff - User wants feedback on implementation - After implementing a feature (self-review)

Shipping pipeline: build, verify, and create PR. Use when: - User says "ship it", "create PR", "open PR", "ready to merge" - User says "/pr", "/ship", "push and PR" - After implementation is complete and needs to be shipped - User wants to verify and publish changes

Testing workflow - write and run tests. Use when: - User says "test", "write tests", "add tests", "run tests" - User mentions "coverage", "unit test", "integration test" - After implementing a feature (tests should follow) - Before shipping or merging code

TLDR code analysis for token-efficient codebase understanding. AUTO-INVOKE when: - User asks "who calls X", "what affects X", "find implementation" - Need to trace dependencies, call graphs, or data flow - Semantic search for code by meaning (not exact string) - Understanding large codebases efficiently - Before reading any large file (use context first) - Debugging "why is X null/undefined here" - Before refactoring (impact analysis) PREFER TLDR over raw file reads - 95% token savings.

Parallel divergent ideation — framed generators, then a critic scores, clusters, deepens top 3. Triggers "/adhd", "brainstorm", "ideate", "widen the option space", fuzzy design/debugging. Skip lookups, syntax, known-root-cause bugs.

Eight audit modes (codebase — owns "audit the codebase" — docs/process, performance, threat-model, motion, SEO, debt). Triggers "nuclear review", "whole codebase review", "adversarial audit", "fable audit", "correctness audit", "audit the docs", "doc drift", "process audit", "perf audit", "why is it slow", "bundle audit", "threat model", "attack surface", "motion audit", "seo audit", "aeo", "llms.txt", "debt ledger", "shortcut ledger". Single-page CWV routes to /lighthouse.

Autonomous skill-prompt optimization — Karpathy-style mutate/score/keep loop on SKILL.md. Triggers "autoresearch", "optimize skill", "tune", "evolve" a skill, "prompt optimization".

Sync cc-settings with Claude Code upstream (maintainer) or update a local Claude Code/Codex install (user). Triggers "sync with claude code", "changelog sync", "upstream sync" (sync mode); "update cc-settings", "refresh my install" (update mode).

Mid-task rollback points — save/restore state before risky ops (refactors, migrations, destructive edits). For end-of-session save use `/handoff`. Triggers "checkpoint", "snapshot", "before this risky op", "rollback to", "list checkpoints", pre-refactor save.

Claude-Codex bridge for bulk/mechanical execution, independent diff review, or read-only advice. Triggers "use codex", "have codex do", "codex exec", "have codex review", "have codex check", "cross-check with codex", "second opinion from codex", "ask codex".

Audit and prune rules/skills/learnings to prevent context bloat. Triggers "consolidate", "clean up rules", "spa day", "what's redundant", degraded agent perf.

Build CONTEXT.md (domain glossary) + docs/adr/ via interview; keeps agent vocab aligned. Triggers "domain language", "glossary", "ADR", inconsistent terminology.

Generate WCAG-checked type, color, spacing, and dark-mode tokens, or consolidate Tailwind v4 tokens with identical rendering. Triggers "design tokens", "type scale", "color palette", "theme setup", "reduce tokens", "dedupe tokens", "consolidate tokens".

Initialize new Darkroom project from a satus or novus starter (NOT the same as native /init which writes CLAUDE.md). Triggers "new darkroom project", "satus", "novus", "scaffold from starter".

Block Edit/Write outside one session directory to contain debugging or parallel agents. Triggers "freeze edits", "lock editing scope", "restrict edits to", "only edit this folder", "unfreeze".

End-of-session save and next-session resume. Triggers "ending session", "wrapping up", "context window", "running out of context", "done for today" (save mode); "continue where we left off", "pick up where", "last session", "previous work", "resume" (resume mode).

Turn proven session behavior into a skill, rule, profile, AGENTS.md diff, or team learning; artifacts seed /autoresearch evals. Triggers "harvest this workflow", "turn this session into a skill", "preserve this behavior", "model handoff".

Lighthouse audit + improvement loop until targets met. Triggers "lighthouse", "page speed", "improve scores", "LCP", "CLS", "INP", "core web vitals". Repo-wide perf audits ("perf audit") go to /audit performance mode instead.

Expert agent in advice, risk, or weighted-comparison mode. Triggers "what should I", "advice on"; "what could go wrong", "risks", "premortem"; or "compare approaches", "which is better", "trade-off analysis", "tech selection".

CEO/founder-mode plan review — stress-tests a plan through a product lens before committing. Triggers "ceo review", "founder review", "product review", "should we even build this", or before a large architectural commitment.

Pre-implementation planning — interview to clarify scope, then compile into a PRD. Triggers "help me figure out", "vague scope", "define requirements" (discovery phase); "PRD", "requirements document", "product spec", "feature spec", "write requirements" (PRD phase).

GitHub Issues as PLAN.md replacement — agents read/update them. Auto-invoke on issue-linked branches. Triggers "what's the plan", "project status", "sync with github", "close the issue".

Run the machine-verifiable gate (typecheck/test/lint, plus a screenshot for UI) so an agent's diff is review-ready before a human sees it. Triggers "proof of work", "is this review-ready", "prove it is green".

Weekly engineering retrospective with persistent metrics — commit patterns, work sessions, code-quality trends. Triggers "retro", "retrospective", "weekly review", "how was my week", "engineering metrics", "velocity", "what did I ship", "show me my stats".

Batch-review diffs from several agents with per-change re-entry cards and reading diffs (real diff, abridged). Triggers "review batch", "batch review", "what's pending review", "reading diff", or after fanning out several agents.

Promote a team-relevant learning to the shared team-knowledge repo, deduping against existing notes first. Triggers "share this", "promote to the team repo", "add to the knowledge base", or after a gotcha/decision/convention worth team-wide awareness.

Product strategy advice on vision, positioning, and how architecture decisions serve them. Triggers "strategist", "product strategy", "market positioning", "what should we build", "product direction".

First-pass review of a client or unfamiliar repo — sweep for glaring issues, report ranked findings. READ-ONLY on external repos. Triggers "triage this repo", "point out glaring issues", "first pass on this client repo".

Adversarial verification — three competing agents (issue-finder, disprover, judge). Triggers "verify", "double check", "are you sure", "poke holes"; pre-prod, post-critical-fix.

Rewrite an accreted patch into its intended end-state by removing unused compatibility layers, flags, fallbacks, and wrappers. Triggers "zero tech debt", "rewrite as if from scratch", "delete the compat layer", "kill the legacy path", "too many flags".