Mat Defect Energy Dft
learningmatter-mit/AtomisticSkills
Calculate charged defect formation energies and transition level diagrams using pymatgen-analysis-defects and atomate2 VASP workflows.
Cross-solver operating discipline for sim-cli workflows — tool choice, input classification, acceptance semantics, and escalation rules that apply across solvers.
Install Sim CLI
$ npx skills add svd-ai-lab/sim-cli --skill sim-cli -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install svd-ai-lab/sim-cli sim-cli --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/svd-ai-lab/sim-cli.git skills-src && mkdir -p .claude/skills && cp -r skills-src/src/sim/_skills/sim-cli .claude/skills/sim-cli && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "sim-cli" agent skill from https://github.com/svd-ai-lab/sim-cli/tree/main/src/sim/_skills/sim-cli into .claude/skills/sim-cli/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "sim-cli", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/svd-ai-lab/sim-cli/tree/main/src/sim/_skills/sim-cliType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add svd-ai-lab/sim-cli --skill sim-cli -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install svd-ai-lab/sim-cli sim-cli --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/svd-ai-lab/sim-cli.git skills-src && mkdir -p .agents/skills && cp -r skills-src/src/sim/_skills/sim-cli .agents/skills/sim-cli && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "sim-cli" agent skill from https://github.com/svd-ai-lab/sim-cli/tree/main/src/sim/_skills/sim-cli into .agents/skills/sim-cli/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "sim-cli", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add svd-ai-lab/sim-cli --skill sim-cli -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install svd-ai-lab/sim-cli sim-cli --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/svd-ai-lab/sim-cli.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/src/sim/_skills/sim-cli .cursor/skills/sim-cli && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "sim-cli" agent skill from https://github.com/svd-ai-lab/sim-cli/tree/main/src/sim/_skills/sim-cli into .cursor/skills/sim-cli/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "sim-cli", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/svd-ai-lab/sim-cli.git --path src/sim/_skills/sim-cli--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add svd-ai-lab/sim-cli --skill sim-cli -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install svd-ai-lab/sim-cli sim-cli --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/svd-ai-lab/sim-cli.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/src/sim/_skills/sim-cli .gemini/skills/sim-cli && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "sim-cli" agent skill from https://github.com/svd-ai-lab/sim-cli/tree/main/src/sim/_skills/sim-cli into .gemini/skills/sim-cli/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "sim-cli", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install svd-ai-lab/sim-cli sim-cliInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add svd-ai-lab/sim-cli --skill sim-cli -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/svd-ai-lab/sim-cli.git skills-src && mkdir -p .github/skills && cp -r skills-src/src/sim/_skills/sim-cli .github/skills/sim-cli && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "sim-cli" agent skill from https://github.com/svd-ai-lab/sim-cli/tree/main/src/sim/_skills/sim-cli into .github/skills/sim-cli/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "sim-cli", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add svd-ai-lab/sim-cli --skill sim-cli -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install svd-ai-lab/sim-cli sim-cli --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/svd-ai-lab/sim-cli.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/src/sim/_skills/sim-cli .opencode/skills/sim-cli && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "sim-cli" agent skill from https://github.com/svd-ai-lab/sim-cli/tree/main/src/sim/_skills/sim-cli into .opencode/skills/sim-cli/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "sim-cli", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
sim-cli- GitHub stars
- 231
- Token cost
- ~2.1k tokens
- SKILL.md length
- 1,128 words
- Files
- 1
- Skills in repo
- 2
- Repo updated
- First seen
- Licence
- Apache-2.0
At a glance
Cross-solver operating discipline for sim-cli workflows — tool choice, input classification, acceptance semantics, and escalation rules that apply across solvers.
- Works in 5 steps: Never invent Category A defaults.… → Step-0 version/profile probe is… → Acceptance ≠ exit code. A sim run / sim… → …
- Tasks that involve Physical and earth sciences
- SKILL.md covers Execution models, Hard constraints (shared…, Input classification and First causal failure gate, plus 2 more sections
- Calls ssh
What it does
Sim CLI is an agent skill from svd-ai-lab/sim-cli. Cross-solver operating discipline for sim-cli workflows — tool choice, input classification, acceptance semantics, and escalation rules that apply across solvers. Use alongside the solver's own plugin skill, which is self-contained for solver-specific work; this skill carries only the shared rules.
Its SKILL.md is about 2.1k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It sits in Development, covering Physical and earth sciences. The repository describes itself as: Turn existing CAE files into structured text an agent can use — COMSOL, Abaqus, Fluent, HFSS, Icepak, FloTHERM — plus solver detection, script linting, and live solver sessions… The licence is Apache-2.0.
When your agent uses it
- Tasks that involve Physical and earth sciences
Example prompts
- “/sim-cli”
Workflow steps
5 steps, taken from the first numbered list in SKILL.md.
- Never invent Category A defaults. Physical decisions — geometry,
- Step-0 version/profile probe is mandatory. After sim connect, call
- Acceptance ≠ exit code. A sim run / sim exec / native solver command
- Never silently retry a failed step. Report stderr, stdout, and
- Reference example values are not defaults. Values in any examples/
What it can do on your machine
Read from SKILL.md and the folder at commit b7318cc. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
ssh
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md. Its commands use ssh, which can reach the network depending on how they are called.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Sim CLI loads about 2.1k tokens when it runs. Until then it costs about 77 tokens; SKILL.md has 1,128 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from svd-ai-lab/sim-cli at commit b7318cc, republished under its Apache-2.0 licence (© svd-ai-lab). 1,128 words, ~2,136 tokens.
Download SKILL.mdSave it as
.claude/skills/sim-cli/SKILL.md (or your agent's skills folder).- name
- sim-cli
- description
- Cross-solver operating discipline for sim-cli workflows — tool choice, input classification, acceptance semantics, and escalation rules that apply across solvers. Use alongside the solver's own plugin skill, which is self-contained for solver-specific work; this skill carries only the shared rules.
sim-cli
You are working in a sim-cli-enabled solver workflow. Use the sim
runtime where it adds control and observability, and compose solver-native
tools directly where they are the right primitive. This skill carries the
cross-solver discipline — the rules that hold across sim-cli workflows,
regardless of which solver you drive.
The solver's plugin skill is self-contained for solver-specific work: solver hard constraints, dependency chains, snippets, workflows, SDK/solver notes. It does not depend on loading this skill — but it should stay consistent with the discipline below. If a rule applies to more than one solver, it belongs here, not in a plugin skill.
Execution models
The plugin skill tells you which model the solver uses. Choose the narrowest correct execution primitive; sim-cli is composable, not a universal wrapper.
| Model | Used by | Lifecycle |
|---|---|---|
| Persistent session | Drivers that hold a live process open | sim connect → sim exec × N → sim inspect → sim disconnect |
| sim-wrapped one-shot batch | Plugin wrappers that add linting, profile-aware invocation, parsing, history, or safer execution | sim check → sim run → parse_output/logs → evaluate |
| Native/tool one-shot batch | Ready solver decks/scripts, vendor batch commands, project pipelines, or post-processing tools where native execution is the right primitive | version/profile probe → native command/script → parse artifacts/logs → evaluate |
sim-cli does not try to wrap every solver operation, executable flag,
vendor batch mode, or post-processing script. Use sim when it adds a stable
control plane: solver discovery and profile checks, live sessions, bounded
exec / inspect loops, history/log capture, parsing, screenshots, or shared
agent discipline. If the artifact is already a solver-native deck/script and
the solver's own batch command is the right execution primitive, call that
native command directly instead of forcing it through sim run. Preserve
stdout/stderr, generated files, and acceptance evidence just as carefully.
Use sim run when the plugin skill says it is the canonical one-shot path or
when the wrapper adds real value. Otherwise, keep the sim-cli discipline while
composing the solver's native tools directly.
The loop, any model: classify inputs and get the missing Category A values
from the user (including the acceptance criterion) → choose the execution
primitive → run a Step-0 version/profile probe (sim inspect session.versions
for persistent sessions; sim check <solver>, plugin guidance, or native
--version / license probe for one-shot work) → execute one bounded step at a
time → inspect last.result or parse the native logs/artifacts between steps →
evaluate against the acceptance criterion → sim disconnect if a persistent
session was opened. On any failure, stop and report — do not silently retry.
Hard constraints (shared sim-cli discipline)
- Never invent Category A defaults. Physical decisions — geometry, materials, boundary conditions, the acceptance criterion — must come from the user. "Just use defaults" / "just run it" does not override this; treat it as a missing input and ask.
- Step-0 version/profile probe is mandatory. After
sim connect, callsim inspect session.versionsand use the returnedprofile/active_sdk_layer/active_solver_layerto pick the right files in the plugin skill. Treatsolver_versionas active-runtime evidence only whenversion_sourceisactive_runtime; a detected installation is not proof that the same version was launched. If the runtime version is unreported or has no compatible profile, stop before applying version-specific guidance. Beforesim runor native one-shot execution, run the relevant probe for that path (sim check <solver>, plugin guidance, or the solver's native--version/ license check). If a required profile is empty, unknown, or deprecated — stop. - Acceptance ≠ exit code. A
sim run/sim exec/ native solver command can return success and still be physically wrong. Always validate against an outcome-based, bounded, measurable criterion — not "the solver ran". - Never silently retry a failed step. Report
stderr,stdout, andrun_count/ completion state / relevant artifact state; let the user decide the next move. - Reference example values are not defaults. Values in any
examples/directory describe a specific published test case. Offer them explicitly if useful, but wait for the user's confirmation before adopting them.
Show full SKILL.md (491 more words)Show less
Input classification
Every task starts with: which inputs must the user supply, which may I default, which can I derive from the files in front of me?
| Category | Rule | Examples |
|---|---|---|
| A — physical decisions | Must ask if absent. Non-negotiable. | Geometry, materials, boundary/initial conditions, physics-model choices, the acceptance criterion |
| B — operational | May default — must disclose. Affects runtime/convenience, not what the simulation represents. | --processors, --ui-mode, --workspace, smoke-test iteration counts, log verbosity |
| C — file-derivable | Infer from the actual files via a diagnostic sim exec, solver-native inspection command, or artifact parser — not from a similar example. Confirm if a downstream decision depends on it. | Mesh cell count, boundary names/types, fields present in a result file, material IDs |
Do not start until every Category A field has an explicit value from the user.
First causal failure gate
When a workflow fails, locate the first failed stage before changing the model: environment/license -> launch or attach -> file import -> model setup -> mesh or build -> solve -> postprocess/export. Preserve the first new error, the exact entrypoint, active version/profile, and relevant artifact identity; later errors may only be consequences.
Run the smallest stage-specific probe that can confirm or reject one cause. Do not change physics, boundary conditions, mesh controls, or solver settings to work around an unproven environment, import, or setup failure. Retry only after naming the hypothesis and the evidence expected to change.
Where sim serve runs (Windows session-context foot-gun)
If you reach a remote sim host via sim --host <host> or SIM_HOST, how
the operator started sim serve changes which drivers actually work. This
is purely a Windows concern — Linux and macOS don't isolate display sessions
the same way.
sim serve started from… | Headless / CLI drivers | GUI-capable drivers |
|---|---|---|
| Logged-in Windows desktop (Windows Terminal / RDP / Task Scheduler "run only when user is logged on" + interactive) | ✅ works | ✅ works — windows are visible; gui can find / click / screenshot them |
SSH session (ssh <host> then sim serve …) | ✅ works | ❌ silent breakage — windows launch in a non-interactive session with no display surface; gui finds zero windows, screenshots come back black |
If the host advertises tools: ["gui"] but gui.find(...) returns nothing
for windows you have strong reason to believe exist, do not retry —
surface "the server may have been started from a non-interactive session" and
ask the operator to restart sim serve from a desktop session. The agent
never starts sim serve itself.
See gui/SKILL.md for the full GUI actuation API.
Long-running and multi-stage workflows
Do not wait for or request a dedicated per-task primitive (a resumable-sweep
API, an auto-coupling feature, etc.) when the generic primitives already
suffice. Bounded exec/run calls, session.health / solve-progress
inspection, and the solver's own retained project/session state are enough to
drive a long sweep, a multi-stage build, or a staged multi-physics coupling
yourself, one bounded step at a time. Reach for a new dedicated primitive only
when a real session hits a generic gap (no timeout guard, no liveness check,
no process cleanup) — not to shortcut a specific workflow that a generic loop
already covers.
© svd-ai-lab, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
Just SKILL.md in src/sim/_skills/sim-cli of svd-ai-lab/sim-cli.
Open the folder on GitHubat commit b7318cc
Compare with similar skills
Sim CLI next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Sim CLI this skillsvd-ai-lab/sim-cli | 231 | — | ~2.1k | Automated safety check: Pass | Apache-2.0 | |
| Mat Defect Energy Dftlearningmatter-mit/AtomisticSkills | 176 | — | ~1.6k | Automated safety check: Pass | MIT | |
| Mat Pourbaix Diagramlearningmatter-mit/AtomisticSkills | 176 | — | ~2.6k | Automated safety check: Pass | MIT | |
| Chemgraphargonne-lcf/ChemGraph | 162 | — | ~2.7k | Automated safety check: Pass | Apache-2.0 | |
| Nanobanana Image GenerationLeoYeAI/openclaw-master-skills | 2.2k | — | ~4.5k | Automated safety check: Pass | MIT | |
| Design for Manufacturing Reviewearthtojake/text-to-cad | 18k | — | ~1.4k | Automated safety check: Pass | MIT |
Similar skills
Mat Pourbaix Diagram
learningmatter-mit/AtomisticSkills
Calculate Pourbaix (pH-voltage) diagrams for aqueous electrochemical stability using water-corrected MLIP energies and pymatgen.
Chemgraph
argonne-lcf/ChemGraph
Develop, test, and extend ChemGraph -- an agentic framework for automated molecular simulations using LLMs, LangGraph, ASE, and MCP servers
Nanobanana Image Generation
LeoYeAI/openclaw-master-skills
A skill your agent uses when the user wants to generate or edit images with Google's Nanobanana/Gemini image models using the official Gemini API shape, or when they need publication-style…
Design for Manufacturing Review
earthtojake/text-to-cad
Guided DFM review of a part for sheet metal, CNC machining or injection molding, covering bends, tool access, draft and undercuts, with evidence-first measurement rules.
Cantera Ignition Delay
K-Dense-AI/scientific-agent-skills
Runs Cantera constant-volume or constant-pressure ignition simulations and reports temperature-based ignition delay with mechanism provenance and checks.
More from svd-ai-lab/sim-cli
Gui
svd-ai-lab/sim-cli
Cross-driver GUI actuation for CAE solvers running under sim-cli.
Related
Categories
Questions about Sim CLI
What does Sim CLI do?
Cross-solver operating discipline for sim-cli workflows — tool choice, input classification, acceptance semantics, and escalation rules that apply across solvers. Sim CLI is an agent skill from svd-ai-lab/sim-cli. Cross-solver operating discipline for sim-cli workflows — tool choice, input classification, acceptance semantics, and escalation rules that apply across solvers.
When should I use Sim CLI?
Sim CLI fits situations like: tasks that involve Physical and earth sciences.
How do I install Sim CLI in Claude Code?
Run `npx skills add svd-ai-lab/sim-cli --skill sim-cli -a claude-code`. Or copy the skill folder (src/sim/_skills/sim-cli in svd-ai-lab/sim-cli) into .claude/skills/sim-cli in your project. Claude Code loads it when a task matches its description.
How do I install Sim CLI in Codex?
Run `npx skills add svd-ai-lab/sim-cli --skill sim-cli -a codex`. Or copy the skill folder (src/sim/_skills/sim-cli in svd-ai-lab/sim-cli) into .agents/skills/sim-cli in your project. Codex loads it when a task matches its description.
Can I use Sim CLI in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add svd-ai-lab/sim-cli --skill sim-cli -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/sim-cli, .gemini/skills/sim-cli, .github/skills/sim-cli and .opencode/skills/sim-cli in your project.
What does Sim CLI need to run?
Going by SKILL.md and its folder, Sim CLI needs the command-line tools its instructions call (ssh).
Does Sim CLI access the network?
SKILL.md contains no URLs. Its commands use ssh, which can reach the network depending on how they are called. This is read from the text; nothing was executed.
Is Sim CLI safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Sim CLI use?
Sim CLI is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Sim CLI use?
About 2.1k tokens (SKILL.md is roughly 8.5k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Sim CLI?
Skills that share tags, products or a category with Sim CLI: Mat Defect Energy Dft (learningmatter-mit/AtomisticSkills, 176 stars), Mat Pourbaix Diagram (learningmatter-mit/AtomisticSkills, 176 stars), Chemgraph (argonne-lcf/ChemGraph, 162 stars) and Nanobanana Image Generation (LeoYeAI/openclaw-master-skills, 2.2k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Sim CLI?
svd-ai-lab (a GitHub organization) maintains it in svd-ai-lab/sim-cli, which has 231 GitHub stars. The repository holds 2 skills in this directory. The repository was last updated on September 8, 2026.
Source: svd-ai-lab/sim-cli on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.