Stats
Actions
Tags
From Claude DB
Orchestrates a full database audit: detects the stack, dispatches read-only subagents in parallel by paradigm, merges findings, and computes Design & Integrity and Performance & Scale scores.
How this skill is triggered — by the user, by Claude, or both
Slash command
/claude-db:db-orchestratorThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Coordinates the whole audit. Three phases: **detect → dispatch → synthesize**. Two scores, never blended.
Coordinates the whole audit. Three phases: detect → dispatch → synthesize. Two scores, never blended.
scripts/detect-stack.mjs). Get the stacks array, each with paradigm, engine, orm, platform, source_of_truth, confidence./claude-db:start (wizard) or description mode. Never guess an engine.established; otherwise those checks will return needs_api.For each stack, spawn the read-only auditor subagents in parallel — one message with multiple Task calls — so their verbose intermediate output stays isolated. (context: fork is NOT a Task parameter; parallelism is achieved by issuing several Task calls in a single message.) Pass each the source_of_truth location/contents, the paradigm, the engine/platform, and its assigned modules. Auditors of an inactive paradigm are not dispatched (their modules emit no findings and leave the denominator):
/claude-db:migrate, or audit when a migration dir exists).Each subagent returns a JSON array of findings conforming to schema/finding.schema.json. Auditors are read-only (tools Read, Grep, Glob, Bash, WebFetch only) — the audit can never mutate files. M19/M20 findings keep their module id for provenance but inherit the natural module's category at scoring time.
id (keep the most severe status: fail > warn > needs_api > pass > not_applicable).node scripts/score.mjs --findings <merged.json> --paradigm <paradigm>
design = min over stores, same for performance); name the flooring store (e.g. "Design 58 — floored by redis-cache").references/scoring-model.md):
score.mjs. If a score is capped:true, show both value and uncapped computed and name the sev-5 fail that capped it.needs_api count across both axes. Never present a needs_api check as a pass.severity × magnitude ÷ effort (magnitude banded high|medium|low, never a fabricated %). Each: status, severity, evidence.observed (real DDL/query, secrets redacted), verification.reproduce, recommendation, fixability (auto/proposed/advisory), and expected_impact (axis + confidence + magnitude)./claude-db:fix to apply the safe, reversible migrations — you confirm each change."Alongside the technical report, give a short plain-language summary: what the two scores mean in one sentence each ("your design is solid but it won't hold load yet"), the top 3 things to fix first in non-jargon terms, and what needs_api means ("I'd need a read-only connection to confirm these"). Keep the expert detail; add the human translation — never replace one with the other.
needs_api when a live DB is required. Read-only by default; speculative findings never cap.npx claudepluginhub hainrixz/claude-db --plugin claude-dbProvides UI/UX resources: 50+ styles, color palettes, font pairings, guidelines, charts for web/mobile across React, Next.js, Vue, Svelte, Tailwind, React Native, Flutter. Aids planning, building, reviewing interfaces.
Searches MemPalace before answering questions about past work, people, projects, or prior decisions. Returns verbatim stored content instead of guessing from model memory.