From c35e1a5fd9541614e3a0becee619687b54b129ef Mon Sep 17 00:00:00 2001 From: "A bot of @njzjz" <48687836+njzjz-bot@users.noreply.github.com> Date: Fri, 5 Jun 2026 12:35:48 +0000 Subject: [PATCH 1/7] docs(agent-skills): make install commands copyable Use concrete OpenClaw install examples instead of shell-hostile angle-bracket placeholders, and explain how to change the target agent. Authored by OpenClaw (model: custom-chat-jinzhezeng-group/gpt-5.5) --- doc/agent-skills.md | 85 +++++++++++++++++++++++++++++++++++++++++++++ doc/index.rst | 1 + 2 files changed, 86 insertions(+) create mode 100644 doc/agent-skills.md diff --git a/doc/agent-skills.md b/doc/agent-skills.md new file mode 100644 index 0000000000..81f025d132 --- /dev/null +++ b/doc/agent-skills.md @@ -0,0 +1,85 @@ +# Agent Skills + +DeePMD-kit provides official [Agent +Skills](https://agentskills.io/what-are-skills) that help AI agents run +DeePMD-kit workflows in a reproducible way. These skills capture +project-specific operating knowledge—such as training inputs, model +deployment, LAMMPS integration, and Python inference patterns—so an agent can +turn a high-level request into concrete files, commands, and validation steps. + +The DeePMD-kit skills were initially developed in the +[Computational Chemistry Agent +Skills](https://github.com/jinzhezenggroup/computational-chemistry-agent-skills) +project as part of the work described below. They are now maintained directly +in the DeePMD-kit repository under `skills/`. + +## List of skills + +- `deepmd-train-dpa3`: Train DeePMD-kit models with the DPA3 descriptor and the + PyTorch backend, including input generation, neighbor-selection choices, + training, freezing, and testing. +- `deepmd-finetune-dpa3`: Fine-tune DPA3 models from self-trained checkpoints, + multi-task pre-trained models, or built-in models downloaded by `dp + pretrained download`. +- `deepmd-train-se-e2-a`: Train classical Deep Potential models with the + `se_e2_a` descriptor, including preparation of training JSON files and + post-training validation. +- `deepmd-python-inference`: Run Python and CLI inference with trained or + frozen DeePMD-kit models, including energy, force, virial, descriptor, and + model-deviation workflows. +- `lammps-deepmd`: Prepare, explain, and run LAMMPS simulations with DeePMD-kit + potentials, including common NVE, NVT, and NPT setups. + +## Related reference + +The DeePMD-kit skills were originally developed as part of the following +paper: + +- Mingwei Ding, Chen Huang, Yibo Hu, Yifan Li, Zitian Lu, Xingtai Yu, Duo + Zhang, Wenxi Zhai, Tong Zhu, Qiangqiang Gu, and Jinzhe Zeng. [Automating + Computational Chemistry Workflows via OpenClaw and Domain-Specific + Skills](https://doi.org/10.48550/arXiv.2603.25522). Accepted by *Journal + of Chemical Theory and Computation*, 2026. + +## Install skills + +### If you are a user + +The easiest way is to send this page to your agent and ask it to install the +skills for you. Users usually do not need to perform manual installation. + +### If you are an agent + +If you already have a DeePMD-kit checkout, run this command from the repository +root: + +```bash +npx -y skills add ./skills -a openclaw -y +``` + +If you do not have a checkout, the same skills can also be installed directly +from GitHub: + +```bash +npx -y skills add https://github.com/deepmodeling/deepmd-kit/tree/master/skills \ + -a openclaw -y +``` + +The examples above install the skills for OpenClaw. Replace `openclaw` with +the target agent name when installing for another agent. The GitHub command +lets the skill CLI fetch the repository for you. For large repositories or +slow networks, this can take longer than installing from an existing local +checkout. Refresh or restart the session afterwards so the installed skills +are reloaded. + +## Minimal verification + +Ask the agent to perform a small task that exercises the installed skill +without launching an expensive calculation. For example: + +- “Use the `deepmd-python-inference` skill to write a minimal Python snippet + for loading a frozen DeePMD-kit model and evaluating one frame.” +- “Use the `deepmd-train-dpa3` skill to draft a small DPA3 training input for a + water dataset, but do not start training.” +- “Use the `lammps-deepmd` skill to prepare an NVT LAMMPS input file for a + DeePMD-kit model, and explain each command.” diff --git a/doc/index.rst b/doc/index.rst index 238dc0d25d..d914782247 100644 --- a/doc/index.rst +++ b/doc/index.rst @@ -45,6 +45,7 @@ DeePMD-kit is a package written in Python/C++, designed to minimize the effort r inference/index cli third-party/index + agent-skills nvnmd/index env troubleshooting/index From 74516eeee762bf95cd07968a37b517caaf0d4469 Mon Sep 17 00:00:00 2001 From: "pre-commit-ci[bot]" <66853113+pre-commit-ci[bot]@users.noreply.github.com> Date: Fri, 5 Jun 2026 12:37:04 +0000 Subject: [PATCH 2/7] [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci --- doc/agent-skills.md | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/doc/agent-skills.md b/doc/agent-skills.md index 81f025d132..e6d2183d88 100644 --- a/doc/agent-skills.md +++ b/doc/agent-skills.md @@ -19,8 +19,7 @@ in the DeePMD-kit repository under `skills/`. PyTorch backend, including input generation, neighbor-selection choices, training, freezing, and testing. - `deepmd-finetune-dpa3`: Fine-tune DPA3 models from self-trained checkpoints, - multi-task pre-trained models, or built-in models downloaded by `dp - pretrained download`. + multi-task pre-trained models, or built-in models downloaded by `dp pretrained download`. - `deepmd-train-se-e2-a`: Train classical Deep Potential models with the `se_e2_a` descriptor, including preparation of training JSON files and post-training validation. @@ -62,7 +61,7 @@ from GitHub: ```bash npx -y skills add https://github.com/deepmodeling/deepmd-kit/tree/master/skills \ - -a openclaw -y + -a openclaw -y ``` The examples above install the skills for OpenClaw. Replace `openclaw` with From bdf5dd4167d4043a5d35e5cb1a532a93809438b5 Mon Sep 17 00:00:00 2001 From: "A bot of @njzjz" <48687836+njzjz-bot@users.noreply.github.com> Date: Fri, 5 Jun 2026 13:00:24 +0000 Subject: [PATCH 3/7] docs(agent-skills): address review comments Apply review feedback by keeping Markdown link text together, documenting the Node.js/npm prerequisite, normalizing terminology and US-English wording, and pinning the GitHub install example to a commit for reproducibility. Authored by OpenClaw (model: custom-chat-jinzhezeng-group/gpt-5.5) --- doc/agent-skills.md | 26 +++++++++++++------------- 1 file changed, 13 insertions(+), 13 deletions(-) diff --git a/doc/agent-skills.md b/doc/agent-skills.md index e6d2183d88..f77c51cb6d 100644 --- a/doc/agent-skills.md +++ b/doc/agent-skills.md @@ -1,15 +1,13 @@ # Agent Skills -DeePMD-kit provides official [Agent -Skills](https://agentskills.io/what-are-skills) that help AI agents run +DeePMD-kit provides official [Agent Skills](https://agentskills.io/what-are-skills) that help AI agents run DeePMD-kit workflows in a reproducible way. These skills capture project-specific operating knowledge—such as training inputs, model deployment, LAMMPS integration, and Python inference patterns—so an agent can turn a high-level request into concrete files, commands, and validation steps. The DeePMD-kit skills were initially developed in the -[Computational Chemistry Agent -Skills](https://github.com/jinzhezenggroup/computational-chemistry-agent-skills) +[Computational Chemistry Agent Skills](https://github.com/jinzhezenggroup/computational-chemistry-agent-skills) project as part of the work described below. They are now maintained directly in the DeePMD-kit repository under `skills/`. @@ -19,7 +17,8 @@ in the DeePMD-kit repository under `skills/`. PyTorch backend, including input generation, neighbor-selection choices, training, freezing, and testing. - `deepmd-finetune-dpa3`: Fine-tune DPA3 models from self-trained checkpoints, - multi-task pre-trained models, or built-in models downloaded by `dp pretrained download`. + multi-task pretrained models, or built-in models downloaded by `dp + pretrained download`. - `deepmd-train-se-e2-a`: Train classical Deep Potential models with the `se_e2_a` descriptor, including preparation of training JSON files and post-training validation. @@ -60,16 +59,17 @@ If you do not have a checkout, the same skills can also be installed directly from GitHub: ```bash -npx -y skills add https://github.com/deepmodeling/deepmd-kit/tree/master/skills \ - -a openclaw -y +npx -y skills add https://github.com/deepmodeling/deepmd-kit/tree/dd428dbb15879115e5eed51aad31cce964782c3c/skills \ + -a openclaw -y ``` -The examples above install the skills for OpenClaw. Replace `openclaw` with -the target agent name when installing for another agent. The GitHub command -lets the skill CLI fetch the repository for you. For large repositories or -slow networks, this can take longer than installing from an existing local -checkout. Refresh or restart the session afterwards so the installed skills -are reloaded. +The examples above require Node.js/npm so that `npx` is available, and they +install the skills for OpenClaw. Replace `openclaw` with the target agent name +when installing for another agent. The GitHub command pins the source to a +specific commit for reproducibility and lets the skill CLI fetch the repository +for you. For large repositories or slow networks, this can take longer than +installing from an existing local checkout. Refresh or restart the session +afterward so the installed skills are reloaded. ## Minimal verification From 349e943ecf9d3f5fe7a1d3d1e840f2a8ed23daf5 Mon Sep 17 00:00:00 2001 From: "pre-commit-ci[bot]" <66853113+pre-commit-ci[bot]@users.noreply.github.com> Date: Fri, 5 Jun 2026 13:01:11 +0000 Subject: [PATCH 4/7] [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci --- doc/agent-skills.md | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/doc/agent-skills.md b/doc/agent-skills.md index f77c51cb6d..d52b0edea3 100644 --- a/doc/agent-skills.md +++ b/doc/agent-skills.md @@ -17,8 +17,7 @@ in the DeePMD-kit repository under `skills/`. PyTorch backend, including input generation, neighbor-selection choices, training, freezing, and testing. - `deepmd-finetune-dpa3`: Fine-tune DPA3 models from self-trained checkpoints, - multi-task pretrained models, or built-in models downloaded by `dp - pretrained download`. + multi-task pretrained models, or built-in models downloaded by `dp pretrained download`. - `deepmd-train-se-e2-a`: Train classical Deep Potential models with the `se_e2_a` descriptor, including preparation of training JSON files and post-training validation. @@ -60,7 +59,7 @@ from GitHub: ```bash npx -y skills add https://github.com/deepmodeling/deepmd-kit/tree/dd428dbb15879115e5eed51aad31cce964782c3c/skills \ - -a openclaw -y + -a openclaw -y ``` The examples above require Node.js/npm so that `npx` is available, and they From 872d9ac5d26318157545b59729ac3113430dd87d Mon Sep 17 00:00:00 2001 From: "A bot of @njzjz" <48687836+njzjz-bot@users.noreply.github.com> Date: Fri, 5 Jun 2026 13:04:02 +0000 Subject: [PATCH 5/7] docs(agent-skills): use master for GitHub install example Keep the remote installation example tracking the maintained master branch instead of pinning it to a commit, so users install the latest skills maintained in the main repository. Authored by OpenClaw (model: custom-chat-jinzhezeng-group/gpt-5.5) --- doc/agent-skills.md | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/doc/agent-skills.md b/doc/agent-skills.md index d52b0edea3..944fd952dd 100644 --- a/doc/agent-skills.md +++ b/doc/agent-skills.md @@ -17,7 +17,8 @@ in the DeePMD-kit repository under `skills/`. PyTorch backend, including input generation, neighbor-selection choices, training, freezing, and testing. - `deepmd-finetune-dpa3`: Fine-tune DPA3 models from self-trained checkpoints, - multi-task pretrained models, or built-in models downloaded by `dp pretrained download`. + multi-task pretrained models, or built-in models downloaded by `dp + pretrained download`. - `deepmd-train-se-e2-a`: Train classical Deep Potential models with the `se_e2_a` descriptor, including preparation of training JSON files and post-training validation. @@ -58,17 +59,16 @@ If you do not have a checkout, the same skills can also be installed directly from GitHub: ```bash -npx -y skills add https://github.com/deepmodeling/deepmd-kit/tree/dd428dbb15879115e5eed51aad31cce964782c3c/skills \ - -a openclaw -y +npx -y skills add https://github.com/deepmodeling/deepmd-kit/tree/master/skills \ + -a openclaw -y ``` The examples above require Node.js/npm so that `npx` is available, and they install the skills for OpenClaw. Replace `openclaw` with the target agent name -when installing for another agent. The GitHub command pins the source to a -specific commit for reproducibility and lets the skill CLI fetch the repository -for you. For large repositories or slow networks, this can take longer than -installing from an existing local checkout. Refresh or restart the session -afterward so the installed skills are reloaded. +when installing for another agent. The GitHub command lets the skill CLI fetch +the repository for you. For large repositories or slow networks, this can take +longer than installing from an existing local checkout. Refresh or restart the +session afterward so the installed skills are reloaded. ## Minimal verification From 61789c97eef8fefa0bb520f960a94d2e77ed1921 Mon Sep 17 00:00:00 2001 From: "pre-commit-ci[bot]" <66853113+pre-commit-ci[bot]@users.noreply.github.com> Date: Fri, 5 Jun 2026 13:04:46 +0000 Subject: [PATCH 6/7] [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci --- doc/agent-skills.md | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/doc/agent-skills.md b/doc/agent-skills.md index 944fd952dd..2413beaf24 100644 --- a/doc/agent-skills.md +++ b/doc/agent-skills.md @@ -17,8 +17,7 @@ in the DeePMD-kit repository under `skills/`. PyTorch backend, including input generation, neighbor-selection choices, training, freezing, and testing. - `deepmd-finetune-dpa3`: Fine-tune DPA3 models from self-trained checkpoints, - multi-task pretrained models, or built-in models downloaded by `dp - pretrained download`. + multi-task pretrained models, or built-in models downloaded by `dp pretrained download`. - `deepmd-train-se-e2-a`: Train classical Deep Potential models with the `se_e2_a` descriptor, including preparation of training JSON files and post-training validation. @@ -60,7 +59,7 @@ from GitHub: ```bash npx -y skills add https://github.com/deepmodeling/deepmd-kit/tree/master/skills \ - -a openclaw -y + -a openclaw -y ``` The examples above require Node.js/npm so that `npx` is available, and they From f8194e3be14d262001a2207ca7957885962e5fa9 Mon Sep 17 00:00:00 2001 From: Jinzhe Zeng Date: Tue, 9 Jun 2026 13:42:34 +0800 Subject: [PATCH 7/7] update paper url Signed-off-by: Jinzhe Zeng --- doc/agent-skills.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/doc/agent-skills.md b/doc/agent-skills.md index 2413beaf24..a78631d445 100644 --- a/doc/agent-skills.md +++ b/doc/agent-skills.md @@ -35,7 +35,7 @@ paper: - Mingwei Ding, Chen Huang, Yibo Hu, Yifan Li, Zitian Lu, Xingtai Yu, Duo Zhang, Wenxi Zhai, Tong Zhu, Qiangqiang Gu, and Jinzhe Zeng. [Automating Computational Chemistry Workflows via OpenClaw and Domain-Specific - Skills](https://doi.org/10.48550/arXiv.2603.25522). Accepted by *Journal + Skills](https://doi.org/10.1021/acs.jctc.6c00622). *Journal of Chemical Theory and Computation*, 2026. ## Install skills