Computational-chemistry-agent-skills relax

Prepare GPAW geometry-relaxation task inputs/scripts from a user-provided structure and essential optimization settings. Use when the user needs structure optimization with explicit optimizer and force-convergence policies.

install

source · Clone the upstream repo

git clone https://github.com/jinzhezenggroup/computational-chemistry-agent-skills

Claude Code · Install into ~/.claude/skills/

T=$(mktemp -d) && git clone --depth=1 https://github.com/jinzhezenggroup/computational-chemistry-agent-skills "$T" && mkdir -p ~/.claude/skills && cp -r "$T/quantum-chemistry/dft-gpaw/relax" ~/.claude/skills/jinzhezenggroup-computational-chemistry-agent-skills-relax-a87e9a && rm -rf "$T"

manifest: quantum-chemistry/dft-gpaw/relax/SKILL.md

source content

GPAW Relaxation (Subskill)

Scope

This skill prepares relaxation tasks only.

It should generate:

structure input
GPAW+ASE optimization script
explicit force/step convergence controls

It should not submit or execute jobs.

Relax intent is mandatory

Before producing the script, classify intent:

ion-only relaxation
cell+ion relaxation (with chosen ASE filter/policy)
low-dimensional/slab-oriented constraints

If intent is ambiguous, ask for clarification.

Must provide

structure input
XC and calculator mode settings
optimizer choice (for example BFGS/FIRE)
force threshold (
```
fmax
```
)
step/trajectory/log policy

Usually should be explicit

stress/cell relaxation policy
constraints for fixed atoms/layers
k-point and smearing settings

Expected output

relaxation-task script/input layout
explicit relaxation-policy rationale
settings summary and assumptions
unresolved choices for user confirmation
handoff note to
```
dpdisp-submit
```
if execution is requested