Menú
Use when a Make It Real installation, slash command, hook, dashboard, current-run, or Claude Code runner environment needs read-only diagnostics.
Instalar con Codex o Claude Copia este prompt, pégalo en Codex, Claude u otro asistente, y deja que revise la página de la skill y la instale por ti.
Basado en la clasificación ocupacional SOC
| name | doctor |
| description | Use when a Make It Real installation, slash command, hook, dashboard, current-run, or Claude Code runner environment needs read-only diagnostics. |
Diagnose the local harness without changing implementation state.
The browser dashboard is read-only observability. Doctor may report dashboard paths and missing preview files, but it must not add browser actions or mutate board, Blueprint, approval, evidence, or config state.
makeitreal-engine doctor "${CLAUDE_PROJECT_DIR:-$PWD}" through the plugin binary.checks.config, checks.plugin, checks.currentRun, checks.hooks, checks.dashboard, and checks.claudeBinary.healthy:false as a diagnostic result, not a command failure.supportMatrix to separate environment problems from harness state problems.nextAction exactly when it is a public command or an install/PATH repair instruction.Lead with a pass/warn/fail diagnostic table and the smallest next fix. Keep raw engine fields, JSON envelopes, run ids, hashes, contract ids, and HARNESS codes in an advanced diagnostic note only when the user asks or troubleshooting requires it. Do not lead with raw engine fields.
Keep the report short: healthy state, failing checks, evidence paths, support matrix highlights, and the next safe action.
Use when a feature request needs Make It Real PRD, architecture, responsibility boundaries, contracts, design-pack artifacts, or Kanban work decomposition before implementation.
Use when a feature request needs Make It Real PRD, architecture, responsibility boundaries, contracts, design-pack artifacts, or Kanban work decomposition before implementation.
Use when an approved Make It Real plan should advance into gated Kanban execution, agent orchestration, verification, and live-wiki synchronization.
Use when an approved Make It Real plan should advance into gated Kanban execution, agent orchestration, verification, and live-wiki synchronization.
Use when a Make It Real run needs manual verification, blocked-gate diagnosis, contract checks, evidence review, or Done-readiness inspection.
Use when a Claude Code project needs Make It Real initialization, hook setup, run-state wiring, or first-time harness configuration before planning or launch.