ArtMin96

Use when the user describes a multi-step feature or refactor and you need to decide whether to handle it solo, dispatch a parallel `/fan-out`, or run a `/worktree-team` planner→generator→reviewer pipeline. Outputs a routing recommendation with the reasoning behind it.

Use whenever the user asks to apply the same operation to many independent files or components — bulk migration, batch refactor across controllers, parallel exploration of unrelated subsystems. Dispatches one subagent per file with a shared prompt template, then collects the results.

Use to toggle the safe-mode flag at .claude/safe-mode. The flag is written automatically by consecutive-failure-guard once tool failures hit FORGE_SAFE_MODE_THRESHOLD (default 5), and block-destructive.sh denies every Bash/Write/Edit while it exists. /safe-mode off clears the flag, logs the exit, and prompts /postmortem; /safe-mode on enters manually; /safe-mode status reports current state.

Run the 5-stage Token Transformation Pipeline (TRAE §5.2.2) over the current session state — Collection, Ranking, Compression, Budgeting, Assembly. Emits a concrete next-action recommendation (/compact, /lean-md, or /progress-log + fresh session) instead of a generic warning.

Audit or scaffold a CLAUDE.md against the Karpathy 4-section structure (Think Before Coding · Simplicity First · Surgical Changes · Goal-Driven Execution). Companion to /lean-md — structure audit, not size trim.

Print every policy enforcement point declared in plugins/diagnostics/registry/policies.json — id, verdict, plugin, hook event, severity, bypass — grouped by verdict. Single discoverable index of what the harness blocks, anchors, nudges, or logs at runtime.

Audit the project against the R.E.S.T. framework (TRAE) — Reliability, Efficiency, Security, Traceability. Reads ledger entries, hook state, and artifact presence across long-session + policy-gateway + context-engine + traces. Single PASS/WARN/FAIL table. Cross-cut meta-surface.

Read the SessionStart timing log written by the diagnostics time-hook wrapper and report per-hook duration plus a cold-vs-warm split across recent sessions. Surfaces which plugin's bootstrap is dominating session-open latency.

Work ONE ticket from a .scratch/<feature>/issues/ local-tracker backlog end to end, in a single command, so each implementation session is one step instead of many. Runs the Soloist start-of-session protocol, picks the next open+unblocked ticket (or the one you name), claims it, implements it test-first via /implement (tdd → gates → code-review → commit), then updates the ticket Status and PROGRESS.md. Use when the user says 'work the next ticket', '/work-ticket', 'do the next PRD/issue', or names a ticket number to implement. One ticket per invocation.

Detect, fix, and re-verify performance, quality, supply-chain, and runtime issues across the whole Soloist app — the Rust core + Tauri adapters and the React/TS frontend — using the project's own detection tools. Use when the user wants to optimize, speed up, audit, health-check, or "find and fix anything broken or slow" in Soloist. Runs the real gates (cargo-deny, clippy, the soak leak-gate, bundle/bloat measurement) and delegates React component issues to the react-doctor skill, then applies fixes and re-runs each gate to confirm green. It changes code, but stays inside the architecture and the locked non-changes — it measures before optimizing, never weakens a test to pass, accepts an unfixable advisory only with a written reason, and never touches PROGRESS.md unless asked.

Cut and ship a Soloist release. Reads what merged since the last release, decides the semantic-version bump, syncs the version across the three manifests, tags, and pushes so CI builds and publishes the GitHub Release the in-app updater reads from. Use when the user wants to release, ship, publish, or bump the app version.

Comprehensively research, review, and verify a set of Soloist changes — a phase's work, a commit range, a PR branch, or the current uncommitted diff — against the architecture, security, clean-room, and codebase-discipline contracts. Use when the user wants to review, audit, or sanity-check Soloist code before merge. Read-and-verify only — it grounds itself in the plan docs, pins the exact diff, researches official docs before judging (never assumes), runs the real gates, and reports findings by severity. It does NOT build features and does NOT change code unless explicitly told to after the findings. First-class checks span domain boundaries & separation (hexagonal + bounded contexts), duplication / single-source / DRY, correctness & bugs, security gaps, performance & size, comment/clean-room discipline, tests, and docs — and it consults the official Tauri docs plus the project's tauri-* skills before judging any Tauri surface.

Implements a Conductor PRD (Product Requirements Document) end-to-end. Reads the PRD, implements all stories in priority order, writes tests, tracks progress, and logs learnings for the next session. Usage: /implement-prd [number] — e.g., /implement-prd 01. If no number given, auto-detects the next uncompleted PRD.

Develops NativePHP Desktop v2 applications. Activates when working with desktop windows, menu bars, system tray, dialogs, alerts, notifications, global shortcuts, clipboard, child processes, auto-updater, or native OS features; or when the user mentions NativePHP, desktop app, native, window management, tray, system integration, or building/publishing desktop apps.