@researai/deepscientist 1.5.16 → 1.6.0

package/src/skills/science/references/packages/abinit.md

@@ -0,0 +1,80 @@
+# ABINIT Electronic-Structure Suite
+
+## Catalog
+
+- Package id: `abinit`
+- Domains: `quantum_chemistry`, `materials_science`
+- Tags: `abinit`, `dft`, `electronic-structure`, `materials-science`, `dfpt`, `first-principles`
+- Knowledge URL: https://github.com/skilled-scipkg/abinit
+- Source archive URL: https://github.com/skilled-scipkg/abinit/archive/refs/heads/master.zip
+- Upstream project URL: https://github.com/abinit/abinit
+- Homepage: https://www.abinit.org
+- Catalog source: FermiLink skilled-scipkg, commit `93f089a333a43089fb1a08a73c37d05fd6683214`
+
+## When To Consider
+
+ABINIT is a first-principles atomic-scale simulation package for computing electronic, structural, vibrational, and dielectric properties of materials using DFT and related methods.
+
+## DeepScientist Runtime Rule
+
+This card is package knowledge and routing context only. It does not mean the
+solver, Python module, CLI binary, compiled backend, license server, dataset, or
+HPC module is installed in the active environment. Before computed work, use
+`bash_exec(...)` to perform an import, executable, version, and smoke-test check
+appropriate for `abinit`.
+
+## Package Check
+
+For CLI/HPC-oriented environments, check the executable or loaded module before
+running any expensive job:
+
+```bash
+command -v abinit || true
+abinit --version || true
+```
+
+If the package is available only through environment modules, record the module
+state and the exact executable path in `validation/environment/abinit_doctor.json`.
+
+Record the result with `artifact.science(...)` as `science.package_check`. Use
+`status="passed"` only when the environment can run at least a minimal smoke
+path. Use `status="failed"` or `status="blocked"` when the check explains why
+execution cannot proceed.
+
+Generated import or executable names are starting points. If `abinit` uses
+a different Python module, CLI binary, environment module, container, or wrapper
+script, adjust the check before concluding the solver is unavailable.
+
+## Expected Science Nodes
+
+- `science.package_check` for import/executable/version/smoke-test evidence
+- `science.computational_run` for solver execution, simulation, fitting, or numerical computation
+- `science.dataset_analysis` when the task primarily analyzes existing data
+- `science.parameter_sweep` when varying parameters, inputs, models, or solver settings
+- `science.validation_result` for convergence, units, schema, controls, or correctness checks
+- `science.claim` only after evidence paths or related nodes support the claim
+
+## Evidence Path Conventions
+
+- `simulations/inputs/` for generated or selected solver inputs
+- `simulations/logs/` for stdout, stderr, scheduler logs, or solver logs
+- `simulations/outputs/` for structured run outputs
+- `analyses/scripts/`, `analyses/logs/`, and `analyses/outputs/` for dataset analysis
+- `validation/environment/` for package checks
+- `validation/runs/` for convergence, unit, schema, or correctness sidecars
+- `figures/` for derived visualizations
+
+## Validation Checklist
+
+- Record package version, executable path, backend, module state, or container image when relevant.
+- Preserve input files and parameters that define the scientific state.
+- Capture units, coordinate conventions, timestep/mesh/basis/model settings, seeds, and convergence criteria when applicable.
+- Validate output schema and important physical or statistical invariants before recording a computed claim.
+- Link claims to run, analysis, sweep, and validation nodes rather than relying on prose.
+
+## Common Pitfalls
+
+- Do not treat the package card or knowledge URL as runtime availability.
+- Do not weaken solver tolerances, physical models, dataset filters, or convergence criteria to make a run pass unless the change is explicitly part of the scientific question.
+- Do not call a value `computed` unless the corresponding run or analysis happened in the current quest and evidence paths are recorded.
+- Do not copy package knowledge-base material into the quest without preserving its source and license context.

package/src/skills/science/references/packages/acts.md

@@ -0,0 +1,73 @@
+# ACTS Common Tracking Software
+
+## Catalog
+
+- Package id: `acts`
+- Domains: `high_energy_physics`, `workflow_provenance`
+- Tags: `high-energy-physics`, `particle-tracking`, `track-reconstruction`, `detector-geometry`, `simulation`, `cpp`
+- Knowledge URL: https://github.com/skilled-scipkg/acts
+- Source archive URL: https://github.com/skilled-scipkg/acts/archive/refs/heads/main.zip
+- Upstream project URL: https://github.com/acts-project/acts
+- Homepage: https://acts-project.github.io
+- Catalog source: FermiLink skilled-scipkg, commit `93f089a333a43089fb1a08a73c37d05fd6683214`
+
+## When To Consider
+
+ACTS is a modern C++ toolkit for experiment independent charged particle track reconstruction, detector geometry, and fast tracking simulation in high energy physics workflows.
+
+## DeepScientist Runtime Rule
+
+This card is package knowledge and routing context only. It does not mean the
+solver, Python module, CLI binary, compiled backend, license server, dataset, or
+HPC module is installed in the active environment. Before computed work, use
+`bash_exec(...)` to perform an import, executable, version, and smoke-test check
+appropriate for `acts`.
+
+## Package Check
+
+Use a package-specific import, executable, or module check and save the result
+under `validation/environment/acts_doctor.json` before treating the
+runtime as usable.
+
+Record the result with `artifact.science(...)` as `science.package_check`. Use
+`status="passed"` only when the environment can run at least a minimal smoke
+path. Use `status="failed"` or `status="blocked"` when the check explains why
+execution cannot proceed.
+
+Generated import or executable names are starting points. If `acts` uses
+a different Python module, CLI binary, environment module, container, or wrapper
+script, adjust the check before concluding the solver is unavailable.
+
+## Expected Science Nodes
+
+- `science.package_check` for import/executable/version/smoke-test evidence
+- `science.computational_run` for solver execution, simulation, fitting, or numerical computation
+- `science.dataset_analysis` when the task primarily analyzes existing data
+- `science.parameter_sweep` when varying parameters, inputs, models, or solver settings
+- `science.validation_result` for convergence, units, schema, controls, or correctness checks
+- `science.claim` only after evidence paths or related nodes support the claim
+
+## Evidence Path Conventions
+
+- `simulations/inputs/` for generated or selected solver inputs
+- `simulations/logs/` for stdout, stderr, scheduler logs, or solver logs
+- `simulations/outputs/` for structured run outputs
+- `analyses/scripts/`, `analyses/logs/`, and `analyses/outputs/` for dataset analysis
+- `validation/environment/` for package checks
+- `validation/runs/` for convergence, unit, schema, or correctness sidecars
+- `figures/` for derived visualizations
+
+## Validation Checklist
+
+- Record package version, executable path, backend, module state, or container image when relevant.
+- Preserve input files and parameters that define the scientific state.
+- Capture units, coordinate conventions, timestep/mesh/basis/model settings, seeds, and convergence criteria when applicable.
+- Validate output schema and important physical or statistical invariants before recording a computed claim.
+- Link claims to run, analysis, sweep, and validation nodes rather than relying on prose.
+
+## Common Pitfalls
+
+- Do not treat the package card or knowledge URL as runtime availability.
+- Do not weaken solver tolerances, physical models, dataset filters, or convergence criteria to make a run pass unless the change is explicitly part of the scientific question.
+- Do not call a value `computed` unless the corresponding run or analysis happened in the current quest and evidence paths are recorded.
+- Do not copy package knowledge-base material into the quest without preserving its source and license context.

package/src/skills/science/references/packages/aiida-core.md

@@ -0,0 +1,80 @@
+# AiiDA Core Workflow Engine
+
+## Catalog
+
+- Package id: `aiida-core`
+- Domains: `workflow_provenance`
+- Tags: `workflow`, `provenance`, `hpc`, `computational-science`, `reproducibility`, `scientific-automation`
+- Knowledge URL: https://github.com/skilled-scipkg/aiida-core
+- Source archive URL: https://github.com/skilled-scipkg/aiida-core/archive/refs/heads/main.zip
+- Upstream project URL: https://github.com/aiidateam/aiida-core
+- Homepage: https://aiida-core.readthedocs.io
+- Catalog source: FermiLink skilled-scipkg, commit `93f089a333a43089fb1a08a73c37d05fd6683214`
+
+## When To Consider
+
+AiiDA-core is a scientific workflow and provenance platform that automates remote HPC calculations, tracks reproducible process graphs, and integrates simulation-code plugins.
+
+## DeepScientist Runtime Rule
+
+This card is package knowledge and routing context only. It does not mean the
+solver, Python module, CLI binary, compiled backend, license server, dataset, or
+HPC module is installed in the active environment. Before computed work, use
+`bash_exec(...)` to perform an import, executable, version, and smoke-test check
+appropriate for `aiida-core`.
+
+## Package Check
+
+For CLI/HPC-oriented environments, check the executable or loaded module before
+running any expensive job:
+
+```bash
+command -v aiida-core || true
+aiida-core --version || true
+```
+
+If the package is available only through environment modules, record the module
+state and the exact executable path in `validation/environment/aiida-core_doctor.json`.
+
+Record the result with `artifact.science(...)` as `science.package_check`. Use
+`status="passed"` only when the environment can run at least a minimal smoke
+path. Use `status="failed"` or `status="blocked"` when the check explains why
+execution cannot proceed.
+
+Generated import or executable names are starting points. If `aiida-core` uses
+a different Python module, CLI binary, environment module, container, or wrapper
+script, adjust the check before concluding the solver is unavailable.
+
+## Expected Science Nodes
+
+- `science.package_check` for import/executable/version/smoke-test evidence
+- `science.computational_run` for solver execution, simulation, fitting, or numerical computation
+- `science.dataset_analysis` when the task primarily analyzes existing data
+- `science.parameter_sweep` when varying parameters, inputs, models, or solver settings
+- `science.validation_result` for convergence, units, schema, controls, or correctness checks
+- `science.claim` only after evidence paths or related nodes support the claim
+
+## Evidence Path Conventions
+
+- `simulations/inputs/` for generated or selected solver inputs
+- `simulations/logs/` for stdout, stderr, scheduler logs, or solver logs
+- `simulations/outputs/` for structured run outputs
+- `analyses/scripts/`, `analyses/logs/`, and `analyses/outputs/` for dataset analysis
+- `validation/environment/` for package checks
+- `validation/runs/` for convergence, unit, schema, or correctness sidecars
+- `figures/` for derived visualizations
+
+## Validation Checklist
+
+- Record package version, executable path, backend, module state, or container image when relevant.
+- Preserve input files and parameters that define the scientific state.
+- Capture units, coordinate conventions, timestep/mesh/basis/model settings, seeds, and convergence criteria when applicable.
+- Validate output schema and important physical or statistical invariants before recording a computed claim.
+- Link claims to run, analysis, sweep, and validation nodes rather than relying on prose.
+
+## Common Pitfalls
+
+- Do not treat the package card or knowledge URL as runtime availability.
+- Do not weaken solver tolerances, physical models, dataset filters, or convergence criteria to make a run pass unless the change is explicitly part of the scientific question.
+- Do not call a value `computed` unless the corresponding run or analysis happened in the current quest and evidence paths are recorded.
+- Do not copy package knowledge-base material into the quest without preserving its source and license context.

package/src/skills/science/references/packages/alamode.md

@@ -0,0 +1,80 @@
+# ALAMODE Lattice Thermal Transport
+
+## Catalog
+
+- Package id: `alamode`
+- Domains: `materials_science`
+- Tags: `materials-science`, `phonons`, `thermal-transport`, `anharmonicity`, `lattice-dynamics`
+- Knowledge URL: https://github.com/skilled-scipkg/ALAMODE
+- Source archive URL: https://github.com/skilled-scipkg/ALAMODE/archive/refs/heads/develop.zip
+- Upstream project URL: https://github.com/ttadano/ALAMODE
+- Homepage: http://sourceforge.net/projects/alamode
+- Catalog source: FermiLink skilled-scipkg, commit `93f089a333a43089fb1a08a73c37d05fd6683214`
+
+## When To Consider
+
+ALAMODE computes harmonic and anharmonic force constants from first-principles data to analyze phonons, lattice anharmonicity, and lattice thermal conductivity in crystalline and low-dimensional solids.
+
+## DeepScientist Runtime Rule
+
+This card is package knowledge and routing context only. It does not mean the
+solver, Python module, CLI binary, compiled backend, license server, dataset, or
+HPC module is installed in the active environment. Before computed work, use
+`bash_exec(...)` to perform an import, executable, version, and smoke-test check
+appropriate for `alamode`.
+
+## Package Check
+
+For CLI/HPC-oriented environments, check the executable or loaded module before
+running any expensive job:
+
+```bash
+command -v alamode || true
+alamode --version || true
+```
+
+If the package is available only through environment modules, record the module
+state and the exact executable path in `validation/environment/alamode_doctor.json`.
+
+Record the result with `artifact.science(...)` as `science.package_check`. Use
+`status="passed"` only when the environment can run at least a minimal smoke
+path. Use `status="failed"` or `status="blocked"` when the check explains why
+execution cannot proceed.
+
+Generated import or executable names are starting points. If `alamode` uses
+a different Python module, CLI binary, environment module, container, or wrapper
+script, adjust the check before concluding the solver is unavailable.
+
+## Expected Science Nodes
+
+- `science.package_check` for import/executable/version/smoke-test evidence
+- `science.computational_run` for solver execution, simulation, fitting, or numerical computation
+- `science.dataset_analysis` when the task primarily analyzes existing data
+- `science.parameter_sweep` when varying parameters, inputs, models, or solver settings
+- `science.validation_result` for convergence, units, schema, controls, or correctness checks
+- `science.claim` only after evidence paths or related nodes support the claim
+
+## Evidence Path Conventions
+
+- `simulations/inputs/` for generated or selected solver inputs
+- `simulations/logs/` for stdout, stderr, scheduler logs, or solver logs
+- `simulations/outputs/` for structured run outputs
+- `analyses/scripts/`, `analyses/logs/`, and `analyses/outputs/` for dataset analysis
+- `validation/environment/` for package checks
+- `validation/runs/` for convergence, unit, schema, or correctness sidecars
+- `figures/` for derived visualizations
+
+## Validation Checklist
+
+- Record package version, executable path, backend, module state, or container image when relevant.
+- Preserve input files and parameters that define the scientific state.
+- Capture units, coordinate conventions, timestep/mesh/basis/model settings, seeds, and convergence criteria when applicable.
+- Validate output schema and important physical or statistical invariants before recording a computed claim.
+- Link claims to run, analysis, sweep, and validation nodes rather than relying on prose.
+
+## Common Pitfalls
+
+- Do not treat the package card or knowledge URL as runtime availability.
+- Do not weaken solver tolerances, physical models, dataset filters, or convergence criteria to make a run pass unless the change is explicitly part of the scientific question.
+- Do not call a value `computed` unless the corresponding run or analysis happened in the current quest and evidence paths are recorded.
+- Do not copy package knowledge-base material into the quest without preserving its source and license context.

package/src/skills/science/references/packages/amuse.md

@@ -0,0 +1,88 @@
+# AMUSE Astrophysical Simulation Environment
+
+## Catalog
+
+- Package id: `amuse`
+- Domains: `astronomy_astrophysics`, `computational_fluid_dynamics`, `workflow_provenance`
+- Tags: `amuse`, `astrophysics`, `computational_astronomy`, `nbody`, `hydrodynamics`, `stellar_evolution`
+- Knowledge URL: https://github.com/skilled-scipkg/amuse
+- Source archive URL: https://github.com/skilled-scipkg/amuse/archive/refs/heads/main.zip
+- Upstream project URL: https://github.com/amusecode/amuse
+- Homepage: http://www.amusecode.org
+- Catalog source: FermiLink skilled-scipkg, commit `93f089a333a43089fb1a08a73c37d05fd6683214`
+
+## When To Consider
+
+AMUSE is a Python-driven framework that couples specialized community solvers to build multi-physics astrophysical simulations across gravitational dynamics, hydrodynamics, stellar evolution, and other computational astronomy research workflows.
+
+## DeepScientist Runtime Rule
+
+This card is package knowledge and routing context only. It does not mean the
+solver, Python module, CLI binary, compiled backend, license server, dataset, or
+HPC module is installed in the active environment. Before computed work, use
+`bash_exec(...)` to perform an import, executable, version, and smoke-test check
+appropriate for `amuse`.
+
+## Package Check
+
+For Python-facing environments, start with an import/version check and then a
+minimal package-specific smoke test:
+
+```bash
+python - <<'PY'
+import importlib, json, pathlib
+package_id = 'amuse'
+result = {"package_id": package_id, "import": "failed", "version": None, "smoke": "not_run"}
+try:
+    module = importlib.import_module(package_id.replace('-', '_').split('_jl')[0])
+    result["import"] = "passed"
+    result["version"] = getattr(module, "__version__", None)
+except Exception as exc:
+    result["error"] = repr(exc)
+pathlib.Path("validation/environment").mkdir(parents=True, exist_ok=True)
+pathlib.Path(f"validation/environment/{package_id}_doctor.json").write_text(json.dumps(result, indent=2), encoding="utf-8")
+PY
+```
+
+Record the result with `artifact.science(...)` as `science.package_check`. Use
+`status="passed"` only when the environment can run at least a minimal smoke
+path. Use `status="failed"` or `status="blocked"` when the check explains why
+execution cannot proceed.
+
+Generated import or executable names are starting points. If `amuse` uses
+a different Python module, CLI binary, environment module, container, or wrapper
+script, adjust the check before concluding the solver is unavailable.
+
+## Expected Science Nodes
+
+- `science.package_check` for import/executable/version/smoke-test evidence
+- `science.computational_run` for solver execution, simulation, fitting, or numerical computation
+- `science.dataset_analysis` when the task primarily analyzes existing data
+- `science.parameter_sweep` when varying parameters, inputs, models, or solver settings
+- `science.validation_result` for convergence, units, schema, controls, or correctness checks
+- `science.claim` only after evidence paths or related nodes support the claim
+
+## Evidence Path Conventions
+
+- `simulations/inputs/` for generated or selected solver inputs
+- `simulations/logs/` for stdout, stderr, scheduler logs, or solver logs
+- `simulations/outputs/` for structured run outputs
+- `analyses/scripts/`, `analyses/logs/`, and `analyses/outputs/` for dataset analysis
+- `validation/environment/` for package checks
+- `validation/runs/` for convergence, unit, schema, or correctness sidecars
+- `figures/` for derived visualizations
+
+## Validation Checklist
+
+- Record package version, executable path, backend, module state, or container image when relevant.
+- Preserve input files and parameters that define the scientific state.
+- Capture units, coordinate conventions, timestep/mesh/basis/model settings, seeds, and convergence criteria when applicable.
+- Validate output schema and important physical or statistical invariants before recording a computed claim.
+- Link claims to run, analysis, sweep, and validation nodes rather than relying on prose.
+
+## Common Pitfalls
+
+- Do not treat the package card or knowledge URL as runtime availability.
+- Do not weaken solver tolerances, physical models, dataset filters, or convergence criteria to make a run pass unless the change is explicitly part of the scientific question.
+- Do not call a value `computed` unless the corresponding run or analysis happened in the current quest and evidence paths are recorded.
+- Do not copy package knowledge-base material into the quest without preserving its source and license context.

package/src/skills/science/references/packages/anndata.md

@@ -0,0 +1,88 @@
+# AnnData Annotated Data Matrices
+
+## Catalog
+
+- Package id: `anndata`
+- Domains: `bioinformatics`, `workflow_provenance`
+- Tags: `anndata`, `single-cell`, `omics`, `bioinformatics`, `h5ad`
+- Knowledge URL: https://github.com/skilled-scipkg/anndata
+- Source archive URL: https://github.com/skilled-scipkg/anndata/archive/refs/heads/main.zip
+- Upstream project URL: https://github.com/scverse/anndata
+- Homepage: http://anndata.readthedocs.io
+- Catalog source: FermiLink skilled-scipkg, commit `93f089a333a43089fb1a08a73c37d05fd6683214`
+
+## When To Consider
+
+anndata provides an efficient annotated matrix container for single-cell and spatial omics, supporting metadata-rich observations, variables, sparse storage, and on-disk workflows.
+
+## DeepScientist Runtime Rule
+
+This card is package knowledge and routing context only. It does not mean the
+solver, Python module, CLI binary, compiled backend, license server, dataset, or
+HPC module is installed in the active environment. Before computed work, use
+`bash_exec(...)` to perform an import, executable, version, and smoke-test check
+appropriate for `anndata`.
+
+## Package Check
+
+For Python-facing environments, start with an import/version check and then a
+minimal package-specific smoke test:
+
+```bash
+python - <<'PY'
+import importlib, json, pathlib
+package_id = 'anndata'
+result = {"package_id": package_id, "import": "failed", "version": None, "smoke": "not_run"}
+try:
+    module = importlib.import_module(package_id.replace('-', '_').split('_jl')[0])
+    result["import"] = "passed"
+    result["version"] = getattr(module, "__version__", None)
+except Exception as exc:
+    result["error"] = repr(exc)
+pathlib.Path("validation/environment").mkdir(parents=True, exist_ok=True)
+pathlib.Path(f"validation/environment/{package_id}_doctor.json").write_text(json.dumps(result, indent=2), encoding="utf-8")
+PY
+```
+
+Record the result with `artifact.science(...)` as `science.package_check`. Use
+`status="passed"` only when the environment can run at least a minimal smoke
+path. Use `status="failed"` or `status="blocked"` when the check explains why
+execution cannot proceed.
+
+Generated import or executable names are starting points. If `anndata` uses
+a different Python module, CLI binary, environment module, container, or wrapper
+script, adjust the check before concluding the solver is unavailable.
+
+## Expected Science Nodes
+
+- `science.package_check` for import/executable/version/smoke-test evidence
+- `science.computational_run` for solver execution, simulation, fitting, or numerical computation
+- `science.dataset_analysis` when the task primarily analyzes existing data
+- `science.parameter_sweep` when varying parameters, inputs, models, or solver settings
+- `science.validation_result` for convergence, units, schema, controls, or correctness checks
+- `science.claim` only after evidence paths or related nodes support the claim
+
+## Evidence Path Conventions
+
+- `simulations/inputs/` for generated or selected solver inputs
+- `simulations/logs/` for stdout, stderr, scheduler logs, or solver logs
+- `simulations/outputs/` for structured run outputs
+- `analyses/scripts/`, `analyses/logs/`, and `analyses/outputs/` for dataset analysis
+- `validation/environment/` for package checks
+- `validation/runs/` for convergence, unit, schema, or correctness sidecars
+- `figures/` for derived visualizations
+
+## Validation Checklist
+
+- Record package version, executable path, backend, module state, or container image when relevant.
+- Preserve input files and parameters that define the scientific state.
+- Capture units, coordinate conventions, timestep/mesh/basis/model settings, seeds, and convergence criteria when applicable.
+- Validate output schema and important physical or statistical invariants before recording a computed claim.
+- Link claims to run, analysis, sweep, and validation nodes rather than relying on prose.
+
+## Common Pitfalls
+
+- Do not treat the package card or knowledge URL as runtime availability.
+- Do not weaken solver tolerances, physical models, dataset filters, or convergence criteria to make a run pass unless the change is explicitly part of the scientific question.
+- Do not call a value `computed` unless the corresponding run or analysis happened in the current quest and evidence paths are recorded.
+- Do not copy package knowledge-base material into the quest without preserving its source and license context.

package/src/skills/science/references/packages/arbor.md

@@ -0,0 +1,80 @@
+# Arbor Neural Simulation Library
+
+## Catalog
+
+- Package id: `arbor`
+- Domains: `computational_neuroscience`, `workflow_provenance`
+- Tags: `computational-neuroscience`, `neuron-simulation`, `electrophysiology`, `multi-compartment`, `hpc`
+- Knowledge URL: https://github.com/skilled-scipkg/arbor
+- Source archive URL: https://github.com/skilled-scipkg/arbor/archive/refs/heads/master.zip
+- Upstream project URL: https://github.com/arbor-sim/arbor
+- Homepage: https://arbor-sim.org
+- Catalog source: FermiLink skilled-scipkg, commit `93f089a333a43089fb1a08a73c37d05fd6683214`
+
+## When To Consider
+
+Arbor is a high-performance library for simulating morphologically detailed multi-compartment neuron and network models with performance portability across CPU and GPU architectures.
+
+## DeepScientist Runtime Rule
+
+This card is package knowledge and routing context only. It does not mean the
+solver, Python module, CLI binary, compiled backend, license server, dataset, or
+HPC module is installed in the active environment. Before computed work, use
+`bash_exec(...)` to perform an import, executable, version, and smoke-test check
+appropriate for `arbor`.
+
+## Package Check
+
+For CLI/HPC-oriented environments, check the executable or loaded module before
+running any expensive job:
+
+```bash
+command -v arbor || true
+arbor --version || true
+```
+
+If the package is available only through environment modules, record the module
+state and the exact executable path in `validation/environment/arbor_doctor.json`.
+
+Record the result with `artifact.science(...)` as `science.package_check`. Use
+`status="passed"` only when the environment can run at least a minimal smoke
+path. Use `status="failed"` or `status="blocked"` when the check explains why
+execution cannot proceed.
+
+Generated import or executable names are starting points. If `arbor` uses
+a different Python module, CLI binary, environment module, container, or wrapper
+script, adjust the check before concluding the solver is unavailable.
+
+## Expected Science Nodes
+
+- `science.package_check` for import/executable/version/smoke-test evidence
+- `science.computational_run` for solver execution, simulation, fitting, or numerical computation
+- `science.dataset_analysis` when the task primarily analyzes existing data
+- `science.parameter_sweep` when varying parameters, inputs, models, or solver settings
+- `science.validation_result` for convergence, units, schema, controls, or correctness checks
+- `science.claim` only after evidence paths or related nodes support the claim
+
+## Evidence Path Conventions
+
+- `simulations/inputs/` for generated or selected solver inputs
+- `simulations/logs/` for stdout, stderr, scheduler logs, or solver logs
+- `simulations/outputs/` for structured run outputs
+- `analyses/scripts/`, `analyses/logs/`, and `analyses/outputs/` for dataset analysis
+- `validation/environment/` for package checks
+- `validation/runs/` for convergence, unit, schema, or correctness sidecars
+- `figures/` for derived visualizations
+
+## Validation Checklist
+
+- Record package version, executable path, backend, module state, or container image when relevant.
+- Preserve input files and parameters that define the scientific state.
+- Capture units, coordinate conventions, timestep/mesh/basis/model settings, seeds, and convergence criteria when applicable.
+- Validate output schema and important physical or statistical invariants before recording a computed claim.
+- Link claims to run, analysis, sweep, and validation nodes rather than relying on prose.
+
+## Common Pitfalls
+
+- Do not treat the package card or knowledge URL as runtime availability.
+- Do not weaken solver tolerances, physical models, dataset filters, or convergence criteria to make a run pass unless the change is explicitly part of the scientific question.
+- Do not call a value `computed` unless the corresponding run or analysis happened in the current quest and evidence paths are recorded.
+- Do not copy package knowledge-base material into the quest without preserving its source and license context.

package/src/skills/science/references/packages/arc.md

@@ -0,0 +1,73 @@
+# ARC Alkali Rydberg Calculator
+
+## Catalog
+
+- Package id: `arc`
+- Domains: `computational_science`
+- Tags: `atomic-physics`, `rydberg-atoms`, `alkali-atoms`, `divalent-atoms`, `quantum-optics`, `stark-shifts`
+- Knowledge URL: https://github.com/skilled-scipkg/ARC-Alkali-Rydberg-Calculator
+- Source archive URL: https://github.com/skilled-scipkg/ARC-Alkali-Rydberg-Calculator/archive/refs/heads/master.zip
+- Upstream project URL: https://github.com/nikolasibalic/ARC-Alkali-Rydberg-Calculator
+- Homepage: https://atomcalc.org
+- Catalog source: FermiLink skilled-scipkg, commit `93f089a333a43089fb1a08a73c37d05fd6683214`
+
+## When To Consider
+
+Python toolkit for computing properties of highly excited Rydberg states and interactions in alkali and divalent atoms, including level structure, transition strengths, and electric-field Stark calculations.
+
+## DeepScientist Runtime Rule
+
+This card is package knowledge and routing context only. It does not mean the
+solver, Python module, CLI binary, compiled backend, license server, dataset, or
+HPC module is installed in the active environment. Before computed work, use
+`bash_exec(...)` to perform an import, executable, version, and smoke-test check
+appropriate for `arc`.
+
+## Package Check
+
+Use a package-specific import, executable, or module check and save the result
+under `validation/environment/arc_doctor.json` before treating the
+runtime as usable.
+
+Record the result with `artifact.science(...)` as `science.package_check`. Use
+`status="passed"` only when the environment can run at least a minimal smoke
+path. Use `status="failed"` or `status="blocked"` when the check explains why
+execution cannot proceed.
+
+Generated import or executable names are starting points. If `arc` uses
+a different Python module, CLI binary, environment module, container, or wrapper
+script, adjust the check before concluding the solver is unavailable.
+
+## Expected Science Nodes
+
+- `science.package_check` for import/executable/version/smoke-test evidence
+- `science.computational_run` for solver execution, simulation, fitting, or numerical computation
+- `science.dataset_analysis` when the task primarily analyzes existing data
+- `science.parameter_sweep` when varying parameters, inputs, models, or solver settings
+- `science.validation_result` for convergence, units, schema, controls, or correctness checks
+- `science.claim` only after evidence paths or related nodes support the claim
+
+## Evidence Path Conventions
+
+- `simulations/inputs/` for generated or selected solver inputs
+- `simulations/logs/` for stdout, stderr, scheduler logs, or solver logs
+- `simulations/outputs/` for structured run outputs
+- `analyses/scripts/`, `analyses/logs/`, and `analyses/outputs/` for dataset analysis
+- `validation/environment/` for package checks
+- `validation/runs/` for convergence, unit, schema, or correctness sidecars
+- `figures/` for derived visualizations
+
+## Validation Checklist
+
+- Record package version, executable path, backend, module state, or container image when relevant.
+- Preserve input files and parameters that define the scientific state.
+- Capture units, coordinate conventions, timestep/mesh/basis/model settings, seeds, and convergence criteria when applicable.
+- Validate output schema and important physical or statistical invariants before recording a computed claim.
+- Link claims to run, analysis, sweep, and validation nodes rather than relying on prose.
+
+## Common Pitfalls
+
+- Do not treat the package card or knowledge URL as runtime availability.
+- Do not weaken solver tolerances, physical models, dataset filters, or convergence criteria to make a run pass unless the change is explicitly part of the scientific question.
+- Do not call a value `computed` unless the corresponding run or analysis happened in the current quest and evidence paths are recorded.
+- Do not copy package knowledge-base material into the quest without preserving its source and license context.