feat: add Atomizer HQ multi-agent cluster infrastructure

- 8-agent OpenClaw cluster (Manager, Tech-Lead, Secretary, Auditor,
  Optimizer, Study-Builder, NX-Expert, Webster)
- Orchestration engine: orchestrate.py (sync delegation + handoffs)
- Workflow engine: YAML-defined multi-step pipelines
- Agent workspaces: SOUL.md, AGENTS.md, MEMORY.md per agent
- Shared skills: delegate, orchestrate, atomizer-protocols
- Capability registry (AGENTS_REGISTRY.json)
- Cluster management: cluster.sh, systemd template
- All secrets replaced with env var references

hq/skills/atomizer-company/LAC_CRITICAL.md

@@ -0,0 +1,39 @@
+# LAC Critical Lessons — NEVER FORGET
+
+These are hard-won insights from past optimization sessions. Violating any of these will cause failures.
+
+## NX Safety (CRITICAL)
+- **NEVER kill ugraf.exe directly** → use `NXSessionManager.close_nx_if_allowed()`
+- **PowerShell for NX journals** → NEVER use `cmd /c`
+- **Always load `*_i.prt` before `UpdateFemodel()`** → mesh won't update without the idealized part
+- **File chain must be intact:** `.sim → .fem → *_i.prt → .prt` (ALL must be present)
+
+## Optimization (CRITICAL)
+- **CMA-ES doesn't evaluate x0 first** → always call `enqueue_trial(x0)` to evaluate baseline
+- **Surrogate + L-BFGS = DANGEROUS** → gradient descent finds fake optima on surrogate surface
+- **NEVER rewrite `run_optimization.py` from scratch** → ALWAYS copy a working template (V15 NSGA-II is gold standard)
+- **Relative WFE math:** use `extract_relative()` (node-by-node subtraction) → NOT `abs(RMS_a - RMS_b)` (wrong math!)
+
+## File Management (IMPORTANT)
+- **Trial folders:** `trial_NNNN/` — zero-padded, never reused, never overwritten
+- **Always copy working studies** — never modify originals
+- **Output paths must be relative** — no absolute Windows/Linux paths (Syncthing-compatible)
+- **Never delete trial data mid-run** — archive after study is complete
+
+## Algorithm Selection (REFERENCE)
+| Variables | Landscape | Recommended | Notes |
+|-----------|-----------|-------------|-------|
+| < 5 | Smooth | Nelder-Mead or COBYLA | Simple, fast convergence |
+| 5-20 | Noisy | CMA-ES | Robust, population-based |
+| > 20 | Any | Bayesian (Optuna TPE) | Efficient with many variables |
+| Multi-obj | Any | NSGA-II or MOEA/D | Pareto front generation |
+| With surrogate | Expensive eval | GNN surrogate + CMA-ES | Reduce simulation count |
+
+## Common Failures
+| Symptom | Cause | Fix |
+|---------|-------|-----|
+| Mesh not updating | Missing `*_i.prt` load | Load idealized part first |
+| NX crashes on journal | Using `cmd /c` | Switch to PowerShell |
+| Baseline trial missing | CMA-ES skips x0 | Explicitly enqueue baseline |
+| Optimization finds unphysical optimum | Surrogate + gradient | Switch to CMA-ES or add validation |
+| Study can't resume | Absolute paths in script | Use relative paths |

hq/skills/atomizer-company/SKILL.md

@@ -0,0 +1,70 @@
+# Atomizer Company Skill
+
+## Description
+Core company identity, values, and agent directory for Atomizer Engineering Co.
+
+## Company Overview
+
+**Atomizer Engineering Co.** is an AI-powered FEA optimization company.
+- **CEO:** Antoine Letarte — Mechanical engineer, freelancer, FEA/optimization specialist
+- **Platform:** Clawdbot multi-agent system on dedicated Slack workspace
+- **Core business:** Structural optimization using Finite Element Analysis
+- **Infrastructure:** Docker on T420 (agents) + Windows/dalidou (NX/Simcenter)
+
+## Company Values
+1. **Engineering rigor first.** Physics is the boss. No shortcuts on validation.
+2. **Ship quality work.** Good enough for the client means good enough for Antoine's reputation.
+3. **Document everything.** Decisions, reasoning, alternatives considered.
+4. **Communicate clearly.** Say what you mean. No jargon for jargon's sake.
+5. **Respect Antoine's time.** He's one person. Filter, summarize, escalate wisely.
+
+## Agent Directory
+
+### Phase 0 (Active)
+| # | Agent | Emoji | ID | Model | Role |
+|---|-------|-------|----|-------|------|
+| 1 | Manager | 🎯 | manager | Opus 4.6 | Orchestrates, delegates, enforces protocols |
+| 2 | Secretary | 📋 | secretary | Opus 4.6 | CEO interface — filters, summarizes, escalates |
+| 3 | Technical Lead | 🔧 | technical-lead | Opus 4.6 | FEA expert, R&D lead, technical reviews |
+
+### Future Phases
+| # | Agent | Emoji | ID | Phase | Role |
+|---|-------|-------|----|-------|------|
+| 4 | Optimizer | ⚡ | optimizer | 1 | Algorithm selection, strategy |
+| 5 | Study Builder | 🏗️ | study-builder | 1 | Writes run_optimization.py |
+| 6 | Auditor | 🔍 | auditor | 1 | Validates physics, challenges assumptions |
+| 7 | NX Expert | 🖥️ | nx-expert | 2 | NX Nastran/NX Open deep knowledge |
+| 8 | Post-Processor | 📊 | post-processor | 2 | Data analysis, visualization |
+| 9 | Reporter | 📝 | reporter | 2 | Professional PDF reports |
+| 10 | Knowledge Base | 🗄️ | knowledge-base | 2 | CAD docs, FEM knowledge library |
+| 11 | Researcher | 🔬 | researcher | 3 | Literature search, state-of-the-art |
+| 12 | Developer | 💻 | developer | 3 | New tools, framework extensions |
+| 13 | IT Support | 🛠️ | it-support | 3 | Infrastructure, licenses, health |
+
+## Communication Hierarchy
+
+```
+Antoine (CEO)
+  ├── 📋 Secretary (direct interface)
+  └── 🎯 Manager (operations)
+        ├── 🔧 Technical Lead
+        ├── ⚡ Optimizer (Phase 1)
+        ├── 🏗️ Study Builder (Phase 1)
+        ├── 🔍 Auditor (Phase 1)
+        └── ... (Phase 2-3 agents)
+```
+
+## Channel Structure
+- `#hq` — Company-wide coordination (Manager's home)
+- `#secretary` — Antoine's private dashboard
+- `#<client>-<project>` — Per-project channels (created as needed)
+- `#rd-<topic>` — R&D exploration channels
+
+## Approval Gates
+Items requiring CEO sign-off:
+- Final deliverables to clients
+- Major technical decisions
+- Budget/cost implications
+- Anything external-facing
+
+Format: `⚠️ **Needs CEO approval:** [summary + recommendation]`