# lammps-simulation-workflows

> This skill should be used when users ask about simulation workflows in lammps; it prioritizes documentation references and then source inspection only for unresolved details.

- Author: Tao E. Li
- Repository: TEL-Research-Group/lammps
- Version: 20260207223556
- Stars: 0
- Forks: 0
- Last Updated: 2026-02-08
- Source: https://github.com/TEL-Research-Group/lammps
- Web: https://mule.run/skillshub/@@TEL-Research-Group/lammps~lammps-simulation-workflows:20260207223556

---

---
name: lammps-simulation-workflows
description: This skill should be used when users ask about simulation workflows in lammps; it prioritizes documentation references and then source inspection only for unresolved details.
---

# lammps: Simulation Workflows

## High-Signal Playbook
### Route the request
- Use `lammps-inputs-and-modeling` for force-field and topology setup before running.
- Use `lammps-analysis-and-output` for thermo/dump/restart output schema.
- Use `lammps-troubleshooting` when runs diverge, stall, or lose atoms.
- Use `lammps-parallel-hpc` for throughput/scaling changes after workflow is stable.

### Triage questions
- Fresh run or continuation from restart/data?
- Which ensemble phases are needed (minimize, NVT, NPT, NVE)?
- What checkpoint cadence is required for wall-time-limited systems?
- Must temperature/pressure ramps persist across segmented runs?
- Are commands inserted between segments (`run ... every`)?

### Canonical workflow
- Start with `run 0` sanity checks and (if needed) minimization (`doc/src/run.rst`, `doc/src/minimize.rst`).
- Equilibrate with appropriate integrator (`fix nvt`/`fix npt`) before production (`doc/src/fix_nh.rst`).
- Use `start/stop` for consistent ramps across multiple runs (`doc/src/run.rst`).
- Write periodic restart files for recovery (`doc/src/restart.rst`, `doc/src/write_restart.rst`).
- Resume with `read_restart` and re-issue non-stored commands (`doc/src/read_restart.rst`, `doc/src/Howto_restart.rst`).
- For queue-limited jobs, use `run ... upto` to continue to a fixed final step (`doc/src/run.rst`).

### Minimal working examples
- Segmented equilibration and checkpointing (`doc/src/run.rst`, `doc/src/fix_nh.rst`):
```LAMMPS
reset_timestep 0
fix eq all nvt temp 300.0 300.0 $(100.0*dt)
thermo 100
run 20000
unfix eq
fix prod all npt temp 300.0 300.0 $(100.0*dt) iso 1.0 1.0 $(1000.0*dt)
restart 50000 restart.prod.*
run 200000 start 0 stop 200000
```
- Restart continuation pattern (`doc/src/read_restart.rst`):
```LAMMPS
read_restart restart.prod.*
neighbor 2.0 bin
neigh_modify every 1 delay 0 check yes
fix int all nve
run 300000 upto
```

### Pitfalls
- Using `run ... pre no` after changing fixes/dumps/neighbor settings (`doc/src/run.rst`).
- Assuming fixes/variables/dump settings are stored in restart files (`doc/src/read_restart.rst`).
- Expecting `run ... every` with `jump` to break execution; use `fix halt` instead (`doc/src/run.rst`).
- Forgetting `reset_timestep` when continuing from converted data files (`doc/src/Howto_restart.rst`).
- Writing restart too infrequently for queue wall-time limits.

### Convergence and validation checklist
- Verify equilibration plateaus before production statistics are accumulated.
- Check restart/read-restart continuity (thermo continuity at restart step).
- Monitor `Dangerous builds`, pressure drift, and energy drift in control runs.
- Confirm ensemble choice matches target observable (NVE conservation, NPT density relaxation).

## Scope
- Handle questions about simulation setup, execution flow, and runtime controls.
- Keep responses abstract and architectural for large codebases; avoid exhaustive per-function documentation unless requested.

## Primary documentation references
- `doc/src/run.rst`
- `doc/src/minimize.rst`
- `doc/src/fix_nve.rst`
- `doc/src/fix_nh.rst`
- `doc/src/velocity.rst`
- `doc/src/restart.rst`
- `doc/src/write_restart.rst`
- `doc/src/read_restart.rst`
- `doc/src/Howto_restart.rst`
- `doc/src/Run_output.rst`
- `bench/in.chain`
- `examples/melt/in.melt`

## Workflow
- Start with the primary references above.
- If details are missing, inspect `references/doc_map.md` for additional topic documents (generated inventory plus curated anchors).
- Use tutorials/examples as executable usage patterns when available.
- Use tests as behavior or regression references when available.
- If ambiguity remains after docs, inspect `references/source_map.md` and start with the ranked source entry points.
- Cite exact documentation file paths in responses.

## Tutorials and examples
- `examples`

## Test references
- None discovered.

## Optional deeper inspection
- `fortran`
- `lib`
- `python`
- `src`

## Source entry points for unresolved issues
- `src/EFF/fix_temp_rescale_eff.h`
- `src/EFF/fix_temp_rescale_eff.cpp`
- `src/fix_thermal_conductivity.h`
- `src/fix_thermal_conductivity.cpp`
- `src/fix_temp_rescale.h`
- `src/fix_temp_rescale.cpp`
- `src/fix_temp_berendsen.h`
- `src/fix_temp_berendsen.cpp`
- Prefer targeted source search (for example: `rg -n "<symbol_or_keyword>" fortran lib python src`).