메뉴
Prepare CP2K single-point (static) task inputs from a user-provided structure and essential DFT settings. Use when the user needs total-energy/electronic SCF evaluation with explicit CP2K basis/potential and SCF controls.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Route NEP requests to task-specific subskills. NEP (Neuroevolution Potential) is the native machine-learning potential family of the GPUMD ecosystem — analogous to DeePMD-kit for LAMMPS. Use when the user asks for `nep.in`, `train.xyz`, `test.xyz`, NEP training, NEP89 reuse, prediction mode, fine-tuning, dipole / polarizability auxiliary models, or automation via NepTrain / NepTrainKit.
Train a first NEP potential from labeled extxyz data. Use when the user needs `nep.in`, `train.xyz`, `test.xyz`, parameter guidance, loss.out interpretation, or deployment of the resulting `nep.txt` back into GPUMD. NEP is the native machine-learning potential for GPUMD and plays the role that DeePMD plays for LAMMPS.
Route VASP DFT requests to task-specific subskills based on user intent. Use when the user asks for VASP workflows and you must decide between static SCF, relaxation, DOS, or band-structure task preparation. This orchestration skill does not own detailed input generation logic; it dispatches to the correct VASP subskill and enforces consistent handoff to submission skills.
Prepare VASP static SCF input tasks from a user-provided structure and essential DFT settings. Use when the user needs single-point electronic structure/total-energy calculations with INCAR generation, KSPACING-based k-point policy (or explicit KPOINTS on request), and POTCAR mapping instructions.
Route ABACUS requests to task-specific subskills based on user intent. Use when the user asks for any ABACUS DFT calculation and you need to determine whether the task is SCF, relaxation, MD, or electronic analysis.
Prepare ABACUS single-point (static SCF) task inputs from a user-provided structure and essential DFT settings. Use when the user needs total-energy/electronic SCF evaluation with explicit ABACUS INPUT/STRU/KPT generation, pseudopotential + orbital mapping, and basis-type selection (PW or LCAO).
| name | static |
| description | Prepare CP2K single-point (static) task inputs from a user-provided structure and essential DFT settings. Use when the user needs total-energy/electronic SCF evaluation with explicit CP2K basis/potential and SCF controls. |
| compatibility | Requires a user-provided structure, suitable CP2K basis/potential files, and runnable CP2K environment. |
| license | GPL-3.0-only |
| catalog-hidden | true |
| metadata | {"author":"qqgu","version":"0.2.0","repository":"https://github.com/cp2k/cp2k"} |
This skill prepares single-point RUN_TYPE ENERGY tasks only. It generates
the CP2K input file, the basis/potential mapping, and CP2K_DATA_DIR guidance.
It does not submit or execute jobs — hand off to dpdisp-submit.
CELL block)PBE, BLYP, SCAN, PADE/LDA, ...)EPS_SCF, MAX_SCF, OT vs diagonalization)MGRID/CUTOFF in Ry (plane-wave auxiliary grid for GPW)LSD flag for open-shell species| Tag | Role | Recommended |
|---|---|---|
CUTOFF | auxiliary grid, Ry | 400 for MOLOPT-SR-GTH, 600 for light atoms + accurate forces |
REL_CUTOFF | ratio cutoff | 60 (default); raise if grid warning appears |
NGRIDS | multigrid levels | 4 (default) |
EPS_DEFAULT | global QS tolerance | 1.0E-10 production, 1.0E-8 screening |
EPS_SCF | inner SCF stop | 1.0E-6 production |
SCF_GUESS | starting density | ATOMIC (cold start), RESTART (chained runs) |
OT | orbital transformation | Always use for insulators (gapped). DIIS + FULL_SINGLE_INVERSE preconditioner is the robust default. |
OUTER_SCF | outer loop for OT | MAX_SCF = 10, EPS_SCF = 1E-6 |
PRINT_LEVEL | verbosity | LOW (routine), MEDIUM (debug) |
OT rules of thumb:
OT — faster, no diagonalization.&SMEAR and ADDED_MOS.LSD .TRUE. and give a MULTIPLICITY hint.MOLOPT-SR-GTH basis sets are optimized per functional. Pair rules:
| Basis family | Potential family | Functional |
|---|---|---|
DZVP-MOLOPT-SR-GTH, TZV2P-MOLOPT-GTH | GTH-PBE-qN | PBE, PBE0, BLYP, SCAN |
DZVP-GTH-PADE, DZV-GTH-PADE | GTH-PADE-qN | LDA / Pade only |
DZVP-MOLOPT-SR-GTH | GTH-BLYP-qN | BLYP (if strict matching is desired) |
Never mix a PADE pseudopotential with a PBE basis, or vice versa — the valence electron count still works but forces and energies become inconsistent with the functional.
Data path: the distribution ships BASIS_MOLOPT, GTH_POTENTIALS,
BASIS_SET, POTENTIAL under $CP2K_DATA_DIR (typically
$CP2K_PREFIX/share/cp2k/data). Either set CP2K_DATA_DIR in the
environment or give an absolute BASIS_SET_FILE_NAME path in the input.
Anchored on the upstream CP2K regression test
tests/QS/regtest-gpw-1/H2O-noheader.inp with the modern MOLOPT basis
and PBE functional (the CP2K howto:static_calculation reference setup).
Task directory:
H2O_scf/
├── h2o.inp
└── (output: H2O-1.restart, H2O-1_0.wfn, h2o.out)
h2o.inp:
&GLOBAL
PROJECT H2O
PRINT_LEVEL LOW
RUN_TYPE ENERGY
&END GLOBAL
&FORCE_EVAL
METHOD QS
&DFT
BASIS_SET_FILE_NAME BASIS_MOLOPT
POTENTIAL_FILE_NAME GTH_POTENTIALS
&MGRID
CUTOFF 400
REL_CUTOFF 60
NGRIDS 4
&END MGRID
&QS
EPS_DEFAULT 1.0E-10
&END QS
&SCF
EPS_SCF 1.0E-6
MAX_SCF 50
SCF_GUESS ATOMIC
&OT
MINIMIZER DIIS
PRECONDITIONER FULL_SINGLE_INVERSE
&END OT
&OUTER_SCF
EPS_SCF 1.0E-6
MAX_SCF 10
&END OUTER_SCF
&END SCF
&XC
&XC_FUNCTIONAL PBE
&END XC_FUNCTIONAL
&END XC
&END DFT
&SUBSYS
&CELL
ABC 10.0 10.0 10.0
PERIODIC XYZ
&END CELL
&COORD
O 0.000000 0.000000 -0.065587
H 0.000000 -0.757136 0.520545
H 0.000000 0.757136 0.520545
&END COORD
&KIND H
BASIS_SET DZVP-MOLOPT-SR-GTH
POTENTIAL GTH-PBE-q1
&END KIND
&KIND O
BASIS_SET DZVP-MOLOPT-SR-GTH
POTENTIAL GTH-PBE-q6
&END KIND
&END SUBSYS
&END FORCE_EVAL
Run (single-thread SMP binary):
export CP2K_DATA_DIR=$CONDA_PREFIX/share/cp2k/data # or path to your build
cp2k.ssmp -i h2o.inp -o h2o.out
grep 'SCF run converged' h2o.out — must report SCF run converged in N steps.
For the H₂O example above expect ~16 inner steps (OT + DIIS) and
a single outer SCF iteration.grep 'ENERGY| Total' h2o.out — total FORCE_EVAL ( QS ) energy.
For the H₂O example (PBE + DZVP-MOLOPT-SR-GTH, CUTOFF 400, 10 Å cubic
box) the reference value is −17.2196 Hartree ≈ −468.43 eV. A drift
1 mHa usually means CUTOFF is too low or the basis/functional do not match.
grep 'Number of electrons' h2o.out — must equal the GTH valence sum
(H₂O: 2 × 1 + 6 = 8). Mismatches indicate a wrong -qN potential
choice.grep 'REL_CUTOFF' h2o.out and check the printed grid distribution —
if the finest grid holds < 50 % of the Gaussians, raise CUTOFF (not
REL_CUTOFF).OT + metal (small-gap Cu, graphene, ...) fails to converge or converges
to a wrong density. Switch to &SCF / DIAGONALIZATION with &SMEAR and
set ADDED_MOS to 2 × occupied.< 6 Å vacuum)
have significant periodic-image interaction — for dipolar solutes use
POISSON / PERIODIC NONE / PSOLVER MT or grow the box.BASIS_SET_FILE_NAME using a relative name requires CP2K_DATA_DIR to
point at the directory that contains the file. A common error is to
point it at the parent share/cp2k-<version> instead of
share/cp2k/data; the error message prints the failed path literally,
which is the quickest way to diagnose.PBE0, HSE06) need auxiliary fit basis
(FIT3/FIT4 in BASIS_ADMM) and the &AUXILIARY_DENSITY_MATRIX_METHOD
block — out of scope for the plain-GGA static skill..inp file and data-dir guidanceCUTOFF rationale, OT vs DIAG choice)dpdisp-submit