scripts/ contains human-invoked repository utilities. Canonical benchmark
entrypoints live in scripts/benchmarks/. Flat files at scripts/*.py may
exist as compatibility shims when historical docs or old commands still point
there.
benchmarks/: benchmark and baseline entrypoints.calibration/: policy bump and synthetic calibration-loop entrypoints.external/: predictor submission, normalization, consensus, and sponsor handoff.lab/: lab handoff, validation, and pass/fail entrypoints.novelty/: novelty DB refresh, audit, and patent-risk entrypoints.release/: demo, evidence, validation, regeneration, and reproducibility entrypoints.research/: exploratory generation, screening, restoration, and dataset-expansion scripts.waves/: wave-program generation, novelty audit, and panel-selection entrypoints.external_validators/: browser and Node-driven external validation helpers.- top-level
*.py: transitional wrappers or non-benchmark operational scripts.
- Flowchart
flowchart TD
User["Contributor or CI"] --> Script["scripts/*"]
Script --> Bench["scripts/benchmarks/*"]
Script --> Ops["Operational utility"]
Bench --> Package["src/openamp_foundry/benchmark"]
Ops --> Package
- Component Diagram
flowchart LR
Makefile --> Scripts
CLI --> Scripts
Scripts --> BenchmarkPkg["benchmark package"]
Scripts --> ReportsPkg["reports / qc / calibration packages"]
External["external_validators"] --> Scripts
- Sequence Diagram
sequenceDiagram
participant User
participant Wrapper as Flat wrapper
participant Canon as Canonical script
participant Lib as src/openamp_foundry/*
User->>Wrapper: run historical command
Wrapper->>Canon: delegate main()
Canon->>Lib: execute benchmark or utility logic
Lib-->>User: artifact or exit code