External Solver Discovery

Organization

svd-ai-lab

external-solver-discovery

Discover and safely use external ad hoc solver capabilities, third-party skills, plugins, MCP servers, or native APIs when the host app or agent environment does not bundle a first-party solver path. Use for unsupported solver requests such as FDTD, Lumerical, Meep, openEMS, Tidy3D, gprMax, CST Studio Suite, or user-provided external integrations.

Overview

Publisher	svd-ai-lab
Repository	`sim-cli`
Skill name	`external-solver-discovery`
Stars	242
Forks	23
Bundled files	1
License	Apache-2.0
Links	Skill source Repository

Markdown instructions
A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.
Works with any LLM
AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.
1 bundled files
Scripts, templates, and references the model can read while it works. Files are read-only and never executed.
Open source
Published by svd-ai-lab on GitHub. Read the source before you install it.

Installation

Install the External Solver Discovery AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

Open the app and go to Plugins → Skills.
Choose "Install from GitHub".
Paste the skill folder URL below and confirm.
https://github.com/svd-ai-lab/sim-cli/tree/main/src/sim/_skills/external-solver-discovery
Enable the skill in any chat where you want it available.

Store

Installed

Skills

Add skills

Import one or more skills from a public GitHub folder or a local .skill or .zip archive, or SKILL.md file.

From GitHub URLFrom .skill, .zip, or SKILL.md

GitHub repo or folder

https://github.com/svd-ai-lab/sim-cli/tree/main/src/sim/_skills/external-solver-discovery

Points to a single skill folder, or any folder that contains multiple skills. Public repos only.

CancelContinue

Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills

git clone --depth 1 https://github.com/svd-ai-lab/sim-cli.git /tmp/sim-cli
mkdir -p .claude/skills
cp -r /tmp/sim-cli/src/sim/_skills/external-solver-discovery .claude/skills/external-solver-discovery

Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable External Solver Discovery in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads External Solver Discovery on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so External Solver Discovery is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

Loaded only when it is needed
The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.
Switch models mid-chat
Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

External Solver Discovery

Use this skill when a user needs a solver, third-party plugin, MCP server, native API, or cloud service that is not already covered by a bundled solver-specific skill.

This skill is an ad hoc discovery layer. It does not make an external solver supported, installed, licensed, authenticated, or available. It helps the agent find credible local or user-configured routes and report evidence honestly.

Required Protocol

Name the requested capability and distinguish it from bundled solver support. For example, HFSS support is 3D EM support, not FDTD support.
Check the current runtime before network-dependent discovery:
- attached files and workspace extensions
- project package files and scripts
- installed commands, environment variables, registry/app paths, and common application locations
- visible skills, plugins, MCP tools, connector tools, and native APIs
- user-provided licenses, paths, profiles, or account hints
If a third-party skill, plugin, or MCP server is visible, inspect its own manifest, README, SKILL file, tool schema, or help output before using it. Prefer integrations that expose provenance, version, tool list, and logs.
Do not install packages, enable cloud services, consume a license seat, submit jobs, or authenticate accounts without user approval.
Probe the smallest real route first:
- version or import check
- open/list a project without solving
- dry script parse
- tiny example run
- MCP health/tool-list call
Treat command success as transport evidence only. Engineering evidence must come from solver logs, exported data, result files, convergence, field data, S-parameters, or another domain-specific output requested by the user.
If no credible route is available, state exactly what is missing and give the narrowest next step, such as a vendor install path, Python environment, package install, MCP server command, API key, or license check.

Candidate Selection

Prefer the user's already visible capability over installing a new one. Use these tie-breakers:

Project-native files first. A .fsp or .lsf project points to a different route than a Python Meep script.
Local native APIs before GUI automation when they can expose the required state and logs.
GUI automation only when the user needs visual operation or the API cannot expose the required state.
Cloud APIs only when the user confirms account, credentials, data-sensitivity constraints, and job-cost expectations.
sim-cli solver packages when one exists for the solver and standardized checks, session control, or sim plugin doctor diagnostics materially reduce risk.

FDTD Ad Hoc Checklist

The host app or agent environment may not bundle first-party FDTD support. When a user asks for FDTD, Lumerical, Meep, openEMS, Tidy3D, gprMax, or files such as .fsp or .lsf, treat the request as external ad hoc discovery.

Candidate routes to check:

Ansys Lumerical FDTD: Prefer when the user has a licensed install or the workspace contains .fsp / .lsf files. Probe user-provided paths, common install locations, and the vendor Python API only after locating a real install. Do not attempt to install Lumerical or obtain licenses.
Meep: Prefer when the project already has Python or Scheme Meep scripts, or an open-source scriptable FDTD path is acceptable. Probe the project Python/conda/container environment with an import or version check before any run. Ask before installing Meep or creating a new environment.
openEMS: Prefer for RF, microwave, antenna, or PCB-oriented workflows, especially when MATLAB/Octave/Python openEMS scripts or CSXCAD references are present. Probe MATLAB, Octave, Python, and PATH before running examples.
Tidy3D: Prefer only when the user confirms a Tidy3D account/API configuration and cloud submission is acceptable for the project data. Run local validation or no-submit construction before uploading or submitting a paid/cloud job.
gprMax: Prefer for ground-penetrating radar or geophysical FDTD problems. Probe input files for gprMax directives and run a version/import check before any full model.

Common project signals include .fsp, .lsf, .ctl, lumapi, import meep, openEMS, CSXCAD, InitFDTD, RunOpenEMS, tidy3d, TIDY3D, gprMax, #domain, #material, and #waveform.

For any FDTD route, collect evidence from the actual tool or API: version or install path, opened project or parsed script, run log, mesh/result metadata, monitor exports, fields, flux data, S-parameters, HDF5/CSV output, or the domain-specific result requested by the user.

CST Third-Party Reference Checklist

The host app or agent environment may not bundle first-party CST Studio Suite support. The public CAE-Agent-Hub repository can be used as a third-party reference when the user explicitly wants a CST MCP route or CST workflow skill:

CST MCP reference: https://github.com/Cai-aa/CAE-Agent-Hub/tree/main/MCP/CST
CST workflow skill reference: https://github.com/Cai-aa/CAE-Agent-Hub/tree/main/Skill/CST/cst-simulation-workflow

Treat these URLs as discovery pointers, not bundled or audited support. Before using them:

Check local evidence first: .cst projects, CST Studio Suite install paths, CST_INSTALL_ROOT, visible MCP server registrations, visible skill folders, project package files, and user-provided paths or license hints.
If CAE-Agent-Hub is present or the user asks to install it, inspect the repository README, CST MCP README, CST skill SKILL.md, and MCP tool surface from that checkout before running tools.
Do not clone, install dependencies, register the MCP server, open CST, run a solver, or consume a CST license seat without user approval.
Prefer a project-local checkout pinned to a tag or commit when installing. Avoid presenting a floating third-party main checkout as a reproducible integration.
Verify the MCP transport with a health, detect, or tool-list probe before opening projects or invoking CST automation.
Work on a copy of any user .cst project unless the user explicitly requests in-place modification.

For CST work, engineering evidence should come from the real CST route: detected version or install path, opened project metadata, saved project path, solver log, result tree entries, exported S-parameters, Touchstone files, field data, far-field or near-field exports, sweep summaries, or the specific result the user requested.

Reporting

Keep the status crisp:

verified: the runtime exposed the tool/API/MCP and a real probe succeeded.
available but unproven: a likely install or config exists but no probe has run yet.
installable: there is a known package/service route, but it is not present.
blocked: a required install, license, credential, path, or network route is missing.

Never describe an ad hoc external route as bundled or first-party support.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

provenance.json1 KB

Frequently asked questions

What does the External Solver Discovery AI skill do?

Why use External Solver Discovery on TypingMind?

Because you install it once and use it with any model. External Solver Discovery is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install External Solver Discovery in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/svd-ai-lab/sim-cli/tree/main/src/sim/_skills/external-solver-discovery. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use External Solver Discovery?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with External Solver Discovery?

As many as you like. As long as a model supports skills, you can use External Solver Discovery with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the External Solver Discovery AI skill free?

Yes. It is published on GitHub by svd-ai-lab under the Apache-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Documentation Lookup

affaan-m

CommunityPopular

Use up-to-date library and framework docs via Context7 MCP instead of training data. Activates for setup questions, API references, code examples, or when the user names a framework (e.g. React,…

Everything Claude Code

affaan-m

CommunityPopular

Development conventions and patterns for everything-claude-code. JavaScript project with conventional commits.

Discord User Post

Post an approved message as the logged-in Discord user through the Discord desktop app. Use for release announcements or other direct user-authored Discord posts; not for OpenClaw channel sends, bots,…

Test Audit

Invoke whenever writing, changing, reviewing, or sweeping tests. Authoring gate for new tests plus audit workflow for low-value, implementation-coupled, or duplicative tests and the test-only…

Overview

Installation

Install in TypingMind

Install in another agent

Use it in TypingMind

Works with any AI model

Skill instructions

External Solver Discovery

Required Protocol

Candidate Selection

FDTD Ad Hoc Checklist

CST Third-Party Reference Checklist

Reporting

Bundled files

Frequently asked questions

Related AI Skills

Set up your own AI workspace now