agentsclimarketplace

Lammps workflows

Skill chatmaterials/lammps-workflows

Agent skill for LAMMPS simulation setup, force-field checks, restarts, and scheduler scripts.

Install
npx -y skills add chatmaterials/lammps-workflows

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 22 stars22 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Use when the task involves LAMMPS atomistic simulation workflows, including input script creation, minimize, NVT, NPT, and restart setups, force-field and data-file checks, log.lammps review, restart handling, and scheduler scripts.

SKILL.md

3.8 KB, as published. Nobody here has run it

LAMMPS Workflows

This skill handles practical LAMMPS setup, review, and restart tasks. Use it when the request is about LAMMPS, in.lammps, log.lammps, force-field wiring, data files, or thermostat and barostat workflows.

When to use

Use this skill when the request mentions or implies:

  • LAMMPS, in.lammps, log.lammps, read_data, pair_style, pair_coeff, thermo
  • minimization, equilibration, NVT, NPT, restart files, or scheduler scripts
  • classical MD or atomistic simulation setup, not plane-wave DFT

Operating stance

Prioritize missing information in this order:

  1. simulation intent: minimize, NVT, NPT, or restart
  2. units, atom style, and force-field family
  3. structure source: read_data, read_restart, or generated structure
  4. temperature, pressure, timestep, and run length
  5. scheduler and MPI or OpenMP layout

Never silently invent:

  • a force field for a chemistry-sensitive system
  • unit systems or atom styles without saying so
  • whether long-range electrostatics, constraints, or special fixes are required
  • whether a restart file is compatible with changed force-field definitions

Workflow

1. Classify the request

  • Setup: create or edit in.lammps, run scripts, and a stage layout.
  • Review: inspect an existing LAMMPS directory and summarize status.
  • Recovery: diagnose input, force-field, or runtime issues and propose the smallest safe correction.

2. Gather the minimum viable context

Before recommending code-specific changes, establish:

  • the intended force field and any required potential files
  • the units and atom style
  • the structure source and whether the box topology is already valid
  • the target ensemble and physical conditions
  • the scheduler environment

3. Use the bundled helpers

  • scripts/make_lammps_inputs.py Generate conservative minimization, NVT, NPT, or restart input skeletons.
  • scripts/check_lammps_job.py Check a LAMMPS directory for missing input, data, or potential-file references.
  • scripts/summarize_lammps_log.py Summarize log.lammps or a run directory using thermo-block heuristics.
  • scripts/recommend_lammps_recovery.py Turn broken or unstable LAMMPS runs into concrete recovery guidance.
  • scripts/export_status_report.py Export a shareable markdown status report from a LAMMPS working directory.
  • scripts/export_input_suggestions.py Export conservative LAMMPS input suggestion snippets based on detected recovery patterns.

4. Load focused references only when needed

  • LAMMPS workflow guidance: references/lammps.md
  • force-field and data-file risks: references/force-fields.md
  • failure handling: references/failure-modes.md
  • scheduler considerations: references/schedulers.md

5. Deliver an auditable answer

Whenever you recommend a LAMMPS change, include:

  • the intended ensemble and units
  • the assumed force field or potential family
  • any unresolved input-file or chemistry decisions the user still needs to confirm
  • what outputs should be checked after the next run

Guardrails

  • A well-formed in.lammps file is not enough if the force field is wrong.
  • Do not treat minimize, NVT, and NPT as interchangeable; explain why the chosen ensemble matches the goal.
  • If a referenced data or potential file is missing, say so directly.

Quality bar

  • Keep workflow advice tied to the actual ensemble and force-field assumptions.
  • Separate syntax issues from model-quality issues.
  • If the current input still contains placeholders, call that out explicitly.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.