Antoine
3289a76e19
- 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
1.2 KiB
1.2 KiB
MEMORY.md — Auditor Long-Term Memory
Common Engineering Pitfalls (always check for)
- Unit inconsistency — especially at interfaces between tools
- Unconverged mesh — results mean nothing without mesh convergence study
- Over-constrained BCs — artificially stiff, unrealistic stress concentrations
- Missing load cases — thermal, dynamic, fatigue often forgotten
- Wrong material direction — anisotropic materials need proper orientation
- Optimization without baseline — can't measure improvement without reference
- Infeasible "optimal" — constraint violations make the result worthless
LAC-Specific Lessons
- CMA-ES doesn't evaluate x0 → baseline trial must be explicit
- Surrogate + L-BFGS → fake optima on approximate surfaces
- Relative WFE computation → use extract_relative()
- NX process management → NXSessionManager.close_nx_if_allowed()
Audit History
(Track completed reviews and recurring findings)
Company Context
- Atomizer Engineering Co. — AI-powered FEA optimization
- Phase 1 agent — quality gatekeeper
- Reviews plans from Optimizer + code from Study Builder + results from Technical Lead
- Has VETO power on deliverables — only Manager or CEO can override