Content
92%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A dense, well-structured CFD skill body: executable code and commands, a gated multi-step workflow with validation checkpoints, and clean one-level-deep reference navigation. The only slack is minor version/plugin verbosity.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Largely efficient and assumes competence (no padding about what CFD/pseudospectral means), but the version/plugin enumerations and dated 2026-07-23 install detail could be trimmed slightly; time-sensitive material is confined to dedicated sections rather than a deprecated/old-patterns section. | 4 / 5 |
Actionability | Fully executable throughout — versioned `uv` install commands, copy-paste `Simul`/forcing/load parameter blocks, and exact bundled-tool invocations like `python3 scripts/solver_config_validator.py --config config.json` cover the common cases. | 5 / 5 |
Workflow Clarity | A 9-step 'Required workflow' with explicit validation checkpoints ('Stop if ...', 'Only then prepare ... Never submit or launch MPI automatically') plus a 'Scientific acceptance gate' checklist and feedback guidance for restart; the destructive/batch MPI risk is explicitly gated. | 5 / 5 |
Progressive Disclosure | SKILL.md is a clear overview with one-level-deep, well-signaled references — all six referenced files (installation, solvers, simulation_workflow, parameters, output_analysis, advanced_features) exist in ./references/ — and bundled scripts are listed with exact invocations and a consolidated References section. | 5 / 5 |
Total | 19 / 20 Passed |