@researai/deepscientist 1.5.16 → 1.6.0

package/src/skills/science/references/packages/dftk-jl.md

@@ -0,0 +1,80 @@
+# DFTK.jl Density-Functional Toolkit
+
+## Catalog
+
+- Package id: `dftk-jl`
+- Domains: `quantum_chemistry`, `materials_science`
+- Tags: `density-functional-theory`, `electronic-structure`, `plane-wave`, `materials-science`, `julia`
+- Knowledge URL: https://github.com/skilled-scipkg/DFTK.jl
+- Source archive URL: https://github.com/skilled-scipkg/DFTK.jl/archive/refs/heads/master.zip
+- Upstream project URL: https://github.com/JuliaMolSim/DFTK.jl
+- Homepage: https://docs.dftk.org
+- Catalog source: FermiLink skilled-scipkg, commit `93f089a333a43089fb1a08a73c37d05fd6683214`
+
+## When To Consider
+
+Julia package for plane-wave density-functional theory in solid-state simulations, emphasizing simple extensible algorithms for electronic-structure research with MPI parallelization and GPU support.
+
+## 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 `dftk-jl`.
+
+## Package Check
+
+For CLI/HPC-oriented environments, check the executable or loaded module before
+running any expensive job:
+
+```bash
+command -v dftk-jl || true
+dftk-jl --version || true
+```
+
+If the package is available only through environment modules, record the module
+state and the exact executable path in `validation/environment/dftk-jl_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 `dftk-jl` 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/dolfinx.md

@@ -0,0 +1,80 @@
+# DOLFINx
+
+## Catalog
+
+- Package id: `dolfinx`
+- Domains: `computational_fluid_dynamics`, `finite_element_engineering`, `workflow_provenance`
+- Tags: `fenicsx`, `finite-element`, `pde`, `hpc`, `multiphysics`
+- Knowledge URL: https://github.com/skilled-scipkg/dolfinx
+- Source archive URL: https://github.com/skilled-scipkg/dolfinx/archive/refs/heads/main.zip
+- Upstream project URL: https://github.com/FEniCS/dolfinx
+- Homepage: https://fenicsproject.org
+- Catalog source: FermiLink skilled-scipkg, commit `93f089a333a43089fb1a08a73c37d05fd6683214`
+
+## When To Consider
+
+DOLFINx is the next generation FEniCSx environment for parallel finite element solution of partial differential equations through interoperable C++ and Python interfaces.
+
+## 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 `dolfinx`.
+
+## Package Check
+
+For CLI/HPC-oriented environments, check the executable or loaded module before
+running any expensive job:
+
+```bash
+command -v dolfinx || true
+dolfinx --version || true
+```
+
+If the package is available only through environment modules, record the module
+state and the exact executable path in `validation/environment/dolfinx_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 `dolfinx` 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/drake.md

@@ -0,0 +1,73 @@
+# Drake Robotics Toolkit
+
+## Catalog
+
+- Package id: `drake`
+- Domains: `robotics_physics`
+- Tags: `robotics`, `simulation`, `multibody`, `optimization`, `control`, `planning`
+- Knowledge URL: https://github.com/skilled-scipkg/drake
+- Source archive URL: https://github.com/skilled-scipkg/drake/archive/refs/heads/master.zip
+- Upstream project URL: https://github.com/RobotLocomotion/drake
+- Homepage: https://drake.mit.edu
+- Catalog source: FermiLink skilled-scipkg, commit `93f089a333a43089fb1a08a73c37d05fd6683214`
+
+## When To Consider
+
+Drake is an open-source robotics toolbox for model-based design, simulation, planning, control, and verification of complex robotic systems.
+
+## 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 `drake`.
+
+## Package Check
+
+Use a package-specific import, executable, or module check and save the result
+under `validation/environment/drake_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 `drake` 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/dumux.md

@@ -0,0 +1,73 @@
+# DuMux Porous Media Simulator
+
+## Catalog
+
+- Package id: `dumux`
+- Domains: `computational_science`
+- Tags: `porousmedia`, `multiphysics`, `finitevolume`, `flowtransport`, `geoscience`
+- Knowledge URL: https://github.com/skilled-scipkg/dumux
+- Source archive URL: https://github.com/skilled-scipkg/dumux/archive/refs/heads/master.zip
+- Upstream project URL: https://github.com/dumux/dumux
+- Homepage: https://dumux.org/
+- Catalog source: FermiLink skilled-scipkg, commit `93f089a333a43089fb1a08a73c37d05fd6683214`
+
+## When To Consider
+
+DuMux is an open source multiphysics simulation framework for finite volume modeling of flow and transport in porous media, supporting coupled domains, reactive processes, and geoscience applications.
+
+## 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 `dumux`.
+
+## Package Check
+
+Use a package-specific import, executable, or module check and save the result
+under `validation/environment/dumux_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 `dumux` 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/elk.md

@@ -0,0 +1,80 @@
+# Elk All-Electron DFT Code
+
+## Catalog
+
+- Package id: `elk`
+- Domains: `quantum_chemistry`, `materials_science`
+- Tags: `electronic-structure`, `density-functional-theory`, `all-electron`, `fp-lapw`, `materials-science`, `solid-state-physics`
+- Knowledge URL: https://github.com/skilled-scipkg/elk
+- Source archive URL: https://github.com/skilled-scipkg/elk/archive/refs/heads/master.zip
+- Upstream project URL: https://github.com/elk-code/elk
+- Homepage: https://elk.sourceforge.io/
+- Catalog source: FermiLink skilled-scipkg, commit `93f089a333a43089fb1a08a73c37d05fd6683214`
+
+## When To Consider
+
+Elk is an all-electron full-potential linearized augmented-plane-wave code for first-principles electronic-structure calculations in solids using density functional theory 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 `elk`.
+
+## Package Check
+
+For CLI/HPC-oriented environments, check the executable or loaded module before
+running any expensive job:
+
+```bash
+command -v elk || true
+elk --version || true
+```
+
+If the package is available only through environment modules, record the module
+state and the exact executable path in `validation/environment/elk_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 `elk` 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/elmerfem.md

@@ -0,0 +1,80 @@
+# Elmer FEM
+
+## Catalog
+
+- Package id: `elmerfem`
+- Domains: `electromagnetics`, `computational_fluid_dynamics`, `finite_element_engineering`
+- Tags: `finite-element`, `multiphysics`, `pde`, `computational-physics`, `cfd`, `structural-mechanics`
+- Knowledge URL: https://github.com/skilled-scipkg/elmerfem
+- Source archive URL: https://github.com/skilled-scipkg/elmerfem/archive/refs/heads/devel.zip
+- Upstream project URL: https://github.com/ElmerCSC/elmerfem
+- Homepage: http://www.elmerfem.org
+- Catalog source: FermiLink skilled-scipkg, commit `93f089a333a43089fb1a08a73c37d05fd6683214`
+
+## When To Consider
+
+ElmerFEM is an open-source multiphysics finite element simulation suite for solving coupled partial differential equations in engineering and physics, including fluid flow, heat transfer, electromagnetics, and structural mechanics.
+
+## 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 `elmerfem`.
+
+## Package Check
+
+For CLI/HPC-oriented environments, check the executable or loaded module before
+running any expensive job:
+
+```bash
+command -v elmerfem || true
+elmerfem --version || true
+```
+
+If the package is available only through environment modules, record the module
+state and the exact executable path in `validation/environment/elmerfem_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 `elmerfem` 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/enzo-e.md

@@ -0,0 +1,88 @@
+# Enzo-E Exascale Astrophysics Simulator
+
+## Catalog
+
+- Package id: `enzo-e`
+- Domains: `astronomy_astrophysics`, `computational_fluid_dynamics`, `workflow_provenance`
+- Tags: `astrophysics`, `cosmology`, `amr`, `hpc`, `hydrodynamics`, `simulation`
+- Knowledge URL: https://github.com/skilled-scipkg/enzo-e
+- Source archive URL: https://github.com/skilled-scipkg/enzo-e/archive/refs/heads/main.zip
+- Upstream project URL: https://github.com/enzo-project/enzo-e
+- Homepage: https://github.com/enzo-project/enzo-e
+- Catalog source: FermiLink skilled-scipkg, commit `93f089a333a43089fb1a08a73c37d05fd6683214`
+
+## When To Consider
+
+Enzo-E is an exascale-oriented astrophysics and cosmology simulation code built on the Cello AMR framework for massively parallel hydrodynamics and structure-formation studies.
+
+## 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 `enzo-e`.
+
+## 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 = 'enzo-e'
+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 `enzo-e` 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/espresso.md

@@ -0,0 +1,80 @@
+# ESPResSo Soft-Matter Molecular Dynamics
+
+## Catalog
+
+- Package id: `espresso`
+- Domains: `molecular_dynamics`, `computational_fluid_dynamics`, `workflow_provenance`
+- Tags: `molecular-dynamics`, `soft-matter`, `coarse-grained`, `computational-physics`, `mesoscale-simulation`
+- Knowledge URL: https://github.com/skilled-scipkg/espresso
+- Source archive URL: https://github.com/skilled-scipkg/espresso/archive/refs/heads/python.zip
+- Upstream project URL: https://github.com/espressomd/espresso
+- Homepage: https://espressomd.org
+- Catalog source: FermiLink skilled-scipkg, commit `93f089a333a43089fb1a08a73c37d05fd6683214`
+
+## When To Consider
+
+ESPResSo is an extensible molecular dynamics engine for coarse-grained soft-matter systems, supporting advanced electrostatics, hydrodynamics coupling, and flexible Python-driven simulation 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 `espresso`.
+
+## Package Check
+
+For CLI/HPC-oriented environments, check the executable or loaded module before
+running any expensive job:
+
+```bash
+command -v espresso || true
+espresso --version || true
+```
+
+If the package is available only through environment modules, record the module
+state and the exact executable path in `validation/environment/espresso_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 `espresso` 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.