atom-audit-fleet

Act as a senior/staff design system orchestrator dispatching N atom-auditor sub-agents and synthesizing their partials into the canonical `ATOM_AUDIT.md` table. Goal: Audit every atom in `packages/design-system/src/atoms/` against `docs/research/COMPONENT_GUIDELINES.md` in parallel: dispatch N Haiku workers (3-5 atoms per worker), wait for convergence, merge their structured rows into `ATOM_AUDIT.md`. ~10x token reduction vs `atom-audit` single-agent expected when atom count >= 6, because Haiku per-token cost is much lower and the rule checks are mechanically schema-bounded. Mandatory context bootstrap (before any output): <!-- shared:mandatory-context-bootstrap --> - Read these sections in `WORKFLOW_OPERATING_SYSTEM.md` first: - `## LLM execution contract` - `## Editor mode policy` (mode definitions only; the tool mapping table is lazy-loaded in `wos/editor-mode-mappings.md` and needed only for non-Claude-Code tools) - `## Global output contract` (including **Adaptive handoff** and **Mode selection rule**) - `## Cross-cutting workflow guardrails` - **Bootstrap tiers (ADR-0025):** the light-weight commands (`branch-commit`, `what-next`, `where-we-at`, `slice-closure`, `compact-task-memory`) may skip `## Editor mode policy` good-fits lists and `## Cross-cutting workflow guardrails` sequencing heuristics, reading only the mode definitions and the core guardrail rules (routing memory, command-less input triage, official command names, material change, no-op). The full ...