← Back to Skills Marketplace
AutoMD-GROMACS
by
Billwanttobetop
· GitHub ↗
· v5.0.0
· MIT-0
210
Downloads
0
Stars
1
Active Installs
4
Versions
Install in OpenClaw
/install automd-gromacs
Description
AutoMD-GROMACS: AI-friendly molecular dynamics automation for GROMACS with workflow, enhanced sampling, special-system simulation, advanced analysis, and pub...
README (SKILL.md)
AutoMD-GROMACS
AutoMD-GROMACS is an AI-oriented automation toolkit for GROMACS. It packages decision-layer routing, end-to-end simulation workflows, enhanced sampling, special-system simulation, advanced analysis, and publication-ready visualization into a public OpenClaw skill with troubleshooting references.
Project metadata:
- Version: 5.0.0
- Author: Guo Xuan
- Organization: Hong Kong University of Science and Technology (Guangzhou)
- Homepage: https://github.com/Billwanttobetop/automd-gromacs
Scope
- Decision layer: method routing before execution via
method-selector - Core workflow: setup, equilibration, production, preprocessing, utilities
- Enhanced sampling: umbrella, free energy, replica exchange, metadynamics, steered MD, enhanced sampling, accelerated MD
- Special systems: membrane, ligand, coarse-grained, electric field, non-equilibrium, QM/MM
- Validation and analysis: trajectory, binding, property, membrane, scattering, free-energy, protein-focused analyses
- Visualization: publication-ready plotting and structure/trajectory rendering
Quick Start
- Read
references/SKILLS_INDEX.yaml - If the method is not chosen yet, use
scripts/decision/method-selector.py - Run the recommended script under
scripts/ - If something fails, read
references/troubleshoot/\x3Cskill>-errors.md
Example:
python3 scripts/decision/method-selector.py \
--goal "binding free energy from a docked complex" \
--system-type protein-ligand \
--target-observable binding-free-energy \
--pretty
Design
- Decision -> execution -> validation as the public product stack
- Executable workflows over tutorial prose
- Layered disclosure for low token overhead
- Embedded domain knowledge from GROMACS practice
- Auto-repair and troubleshooting guidance by default
Project Info
- Version: 5.0.0
- Based on: GROMACS 2026.1
- Runtime needs:
python3,PyYAML,gmx - License: MIT
- Homepage: https://github.com/Billwanttobetop/automd-gromacs
Get started: read references/SKILLS_INDEX.yaml
Usage Guidance
This package is coherent with an AI-friendly GROMACS automation toolkit and appears to be what it claims. Before running anything: (1) ensure gmx and python3 come from trusted installs (conda-forge or upstream), (2) inspect any scripts that auto-download resources or call pip/git before executing them, (3) avoid running publish/push/remote-upload commands (clawhub, git push, scp) from untrusted hosts or without reviewing authorship, and (4) if you want to test, run scripts in a sandboxed environment (container/VM) and with network access disabled until you validate the code. If you need higher assurance, ask the author for a signed/reproducible release or test-run the scripts on a small dummy system first.
Capability Analysis
Type: OpenClaw Skill
Name: automd-gromacs
Version: 5.0.0
AutoMD-GROMACS is a comprehensive and legitimate automation toolkit for GROMACS molecular dynamics simulations. The bundle includes a decision-layer routing system (method-selector.py), extensive simulation workflows, and analysis modules. While the scripts possess broad capabilities such as automatically installing dependencies via pip/conda and fetching scientific data from external domains (e.g., cgmartini.nl), these actions are strictly aligned with the stated purpose of scientific automation. No evidence of data exfiltration, malicious execution, persistence mechanisms, or prompt injection was found across the code or documentation.
Capability Tags
Capability Assessment
Purpose & Capability
Name/description (AutoMD-GROMACS) match the included scripts, analyses, and docs. Declared required binaries (gmx, python3) and runtime needs (PyYAML) are appropriate for the stated MD workflows. The large set of scripts and documentation aligns with the claimed capabilities (workflows, enhanced sampling, analysis, visualization).
Instruction Scope
SKILL.md and many scripts instruct the agent/user to run local scripts (scripts/*), use gmx and python3, and consult references/*. They also include explicit publish/upload steps (git, clawhub, scp) and automatic downloads (forcefields, martinize) in some scripts — these actions are expected for this domain but have side effects (network access, file writes). The instructions do not request unrelated system secrets or arbitrary file exfiltration, but they do encourage operations that can publish or upload the repository if executed.
Install Mechanism
No opaque binary download URLs are embedded; SKILL.md suggests installing GROMACS via conda-forge or following the official manual, and PyYAML via pip. Where scripts auto-download forcefields or tools, the documentation references known project resources (MARTINI, martinize). There is a minor inconsistency between the registry-level 'no install spec' note and the SKILL.md's install guidance, but the install recommendations themselves are proportional and use mainstream sources.
Credentials
The skill declares no required environment variables or credentials and only needs gmx/python3. Some workflows mention interactive login steps (clawhub login, git push) but do not require secrets be provided to the skill. No unexpected credentials or config paths are requested by the skill metadata or documentation.
Persistence & Privilege
always is false and autonomous invocation is allowed (platform default). The skill does not request persistent platform-wide privileges or to modify other skills' configs. Several included docs give instructions for publishing the repository (git/clawhub) — these are user-driven actions, not automatic privileges embedded in the skill.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install automd-gromacs - After installation, invoke the skill by name or use
/automd-gromacs - Provide required inputs per the skill's parameter spec and get structured output
Version History
v5.0.0
Add decision layer with method-selector, canonical target map, V5 public packaging cleanup, and publishable release workflow.
v4.0.0
Expand AutoMD-GROMACS with enhanced sampling, special-system workflows, advanced analysis, and publication-ready visualization.
v1.0.1
- Added PUBLISH_COMPLETE.md and SKILLHUB_PUBLISH_STATUS.md files.
- Updated documentation in README.md.
- No changes to core workflow or functionality.
v1.0.0
- Major update: Now covers 13 molecular dynamics simulation skills, including system setup, equilibration, production, analysis, free energy, ligand binding, membrane proteins, umbrella sampling, PCA, and automated workflows.
- Introduces a 3-layer documentation structure with quick overview, indexed skills, and executable scripts with troubleshooting guides.
- Built-in auto-repair functions to minimize manual intervention.
- Optimized for token efficiency, achieving 84.7% savings over traditional workflows.
- Provides built-in installation options for GROMACS via conda or manual setup.
Metadata
Frequently Asked Questions
What is AutoMD-GROMACS?
AutoMD-GROMACS: AI-friendly molecular dynamics automation for GROMACS with workflow, enhanced sampling, special-system simulation, advanced analysis, and pub... It is an AI Agent Skill for Claude Code / OpenClaw, with 210 downloads so far.
How do I install AutoMD-GROMACS?
Run "/install automd-gromacs" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is AutoMD-GROMACS free?
Yes, AutoMD-GROMACS is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does AutoMD-GROMACS support?
AutoMD-GROMACS is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created AutoMD-GROMACS?
It is built and maintained by Billwanttobetop (@billwanttobetop); the current version is v5.0.0.
More Skills