Claude Skill

agents

Imported from haoming-luo/agentfem/docs/agents.

LLM Mart · 0 points · 6 views 62 listing impressions 0 install-command copies
Download haoming-luo-agentfem-docs_agents-fc86db6.zip · 25 KB

Install

skills CLI npx skills add https://github.com/haoming-luo/agentfem/tree/main/docs/agents
Claude Code claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install haoming-luo-agentfem@llmmart
Git git clone https://github.com/haoming-luo/agentfem.git

The skills CLI installs just this skill, for any of its supported agents. Claude Code installs the whole haoming-luo/agentfem collection as a plugin from our marketplace. Git is the plain clone.

Skill manifest

AgentFEM skill

The repository ships a standards-compatible AgentFEM skill that teaches coding agents the public workflow, project lifecycle, result contract, scientific validation boundary, and safe extension rules.

The canonical source is skills/agentfem/SKILL.md. It is versioned with the code so that agent behavior can evolve alongside API and project-format changes.

The directory is self-contained: keep SKILL.md, agents/, and references/ together. A skill-aware agent can load it directly from the repository. For a personal Codex installation, copy or link the complete skills/agentfem/ directory to $CODEX_HOME/skills/agentfem/ (or ~/.codex/skills/agentfem/ when CODEX_HOME is unset); copying only SKILL.md omits the scientific and implementation references.

The main skill stays concise and routes detailed material progressively: workflow guidance is read for modeling tasks, while concepts, module ownership, validation, extension rules and specialized scientific notes are loaded only when the task requires them. Installation ends with environment readiness; the skill offers a first example without automatically creating or running it.

What the skill is for

  • discover and validate an AgentFEM installation;
  • create and operate projects through the public workflow;
  • interpret structured failures and simulation results;
  • preserve the distinction between execution and scientific verification;
  • route reusable finite-element functionality to the correct module;
  • upgrade older projects without silently rewriting scientific intent.
  • freeze runtime and scientific-input evidence, run safe local ensembles, and build traceable convergence/loading/response experiments;
  • inspect declared private extensions without automatically executing or installing untrusted packages.

For modeling, start with a relevant example and capability topic, then consult provider options as needed. The installed API is authoritative when online documentation describes newer work. New cases use the canonical vocabulary; compatibility methods remain available for older projects.

The skill complements the machine-readable /agentfem.json manifest and scientific src/agentfem/knowledge/catalog.json; it does not replace the public API or the solver's deterministic checks.

Performance and reproducible delivery

The optional companion skill skills/agentfem-performance/SKILL.md guides performance investigation, fast-path migration and reproducible project delivery. Keep its references/ directory with it. Install the complete skills/agentfem-performance/ directory in a supported skill location using the same copy-or-link approach above. The repository is the versioned source; refresh local copies deliberately when updating, preserving local changes.

It distinguishes unchanged-algorithm comparisons from new-method accuracy studies and hardware tuning. Past worker counts, step counts and timings are not universal requirements. It contains no private meshes or material laws. Experimental design, migration parity and delivery checks have separate references: load the one relevant to the task, not a release checklist for every optimization.

Both skills are development guidance distributed under the repository's Apache-2.0 license, not numerical libraries or runtime dependencies. An ordinary prepared project runs through the installed AgentFEM program without a skill, an AI account or an agent editing its files during the solve. Record the actual runtime and test the documented launch command before claiming that a particular delivery supports it; a custom launcher is not evidence that agentfem run has been accepted for that delivery.

Files (agentfem)
  • acceptance.md 3.9 KB
    Loading…
  • index.md 5.8 KB
    Loading…
  • mcp.md 4 KB
    Loading…
  • scientific_reference.md 44.2 KB
    Loading…
  • skill.md 4 KB
    ---
    title: "AgentFEM workflow skill for coding assistants"
    description: "Use the versioned AgentFEM skill to guide an AI assistant through model setup, execution and result inspection using public interfaces."
    ---
    
    # AgentFEM skill
    
    The repository ships a standards-compatible AgentFEM skill that teaches coding
    agents the public workflow, project lifecycle, result contract, scientific
    validation boundary, and safe extension rules.
    
    The canonical source is
    [`skills/agentfem/SKILL.md`](https://github.com/haoming-luo/agentfem/blob/main/skills/agentfem/SKILL.md).
    It is versioned with the code so that agent behavior can evolve alongside API
    and project-format changes.
    
    The directory is self-contained: keep `SKILL.md`, `agents/`, and `references/`
    together. A skill-aware agent can load it directly from the repository. For a
    personal Codex installation, copy or link the complete `skills/agentfem/`
    directory to `$CODEX_HOME/skills/agentfem/` (or
    `~/.codex/skills/agentfem/` when `CODEX_HOME` is unset); copying only
    `SKILL.md` omits the scientific and implementation references.
    
    The main skill stays concise and routes detailed material progressively:
    workflow guidance is read for modeling tasks, while concepts, module ownership,
    validation, extension rules and specialized scientific notes are loaded only when
    the task requires them. Installation ends with environment readiness; the skill
    offers a first example without automatically creating or running it.
    
    ## What the skill is for
    
    - discover and validate an AgentFEM installation;
    - create and operate projects through the public workflow;
    - interpret structured failures and simulation results;
    - preserve the distinction between execution and scientific verification;
    - route reusable finite-element functionality to the correct module;
    - upgrade older projects without silently rewriting scientific intent.
    - freeze runtime and scientific-input evidence, run safe local ensembles, and
      build traceable convergence/loading/response experiments;
    - inspect declared private extensions without automatically executing or
      installing untrusted packages.
    
    For modeling, start with a relevant example and capability topic, then consult
    provider options as needed. The installed API is authoritative when online
    documentation describes newer work. New cases use the canonical vocabulary;
    compatibility methods remain available for older projects.
    
    The skill complements the machine-readable `/agentfem.json` manifest and
    scientific `src/agentfem/knowledge/catalog.json`; it does not replace the public API or the
    solver's deterministic checks.
    
    ## Performance and reproducible delivery
    
    The optional companion skill
    [`skills/agentfem-performance/SKILL.md`](https://github.com/haoming-luo/agentfem/blob/main/skills/agentfem-performance/SKILL.md)
    guides performance investigation, fast-path migration and reproducible project
    delivery. Keep its `references/` directory with it. Install the complete
    `skills/agentfem-performance/` directory in a supported skill location using
    the same copy-or-link approach above. The repository is the versioned source;
    refresh local copies deliberately when updating, preserving local changes.
    
    It distinguishes unchanged-algorithm comparisons from new-method accuracy
    studies and hardware tuning. Past worker counts, step counts and timings are
    not universal requirements. It contains no private meshes or material laws.
    Experimental design, migration parity and delivery checks have separate references:
    load the one relevant to the task, not a release checklist for every optimization.
    
    Both skills are development guidance distributed under the repository's
    Apache-2.0 license, not numerical libraries or runtime dependencies. An ordinary
    prepared project runs through the installed AgentFEM program without a skill,
    an AI account or an agent editing its files during the solve. Record the actual
    runtime and test the documented launch command before claiming that a particular
    delivery supports it; a custom launcher is not evidence that `agentfem run`
    has been accepted for that delivery.
    

Comments (0)

Sign in to join the conversation.

No comments yet.

Reviews (0)

No reviews yet.