Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add LLK-LL/K-agent --skill lsdyna-kfilegit clone --depth 1 https://github.com/LLK-LL/K-agentWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/llk-ll/k-agent/lsdyna-kfile)<a href="https://agentmods.dev/skills/llk-ll/k-agent/lsdyna-kfile"><img src="https://agentmods.dev/badge/skills/llk-ll/k-agent/lsdyna-kfile.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00106 | $0.03211 |
| Opus 5 | $0.00053 | $0.01605 |
| Sonnet 5 | $0.00021 | $0.00642 |
| Haiku 4.5 | $0.00011 | $0.00321 |
Grade A, and why
lsdyna-kfile scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 139 lines — stays where its author put it; the contents beside it link to each section on GitHub.
LS-DYNA K 文件编写与自动验证
交付物只能是经本机求解器实际计算验证的完整 k 文件和验证报告,不可“写完就交”。
以下把本文件所在目录称为 SKILL_DIR;脚本、知识、协议和模板分别位于 SKILL_DIR/scripts/、
SKILL_DIR/knowledge/、SKILL_DIR/references/、SKILL_DIR/templates/。
1. 路由与环境
- 模板制作模式:用户上传主
.k/.key/.dyn,明确说明其要作为某类模板,并要求抽象、整理、 参数化或沉淀。优先执行references/template-authoring-protocol.md;先运行python "SKILL_DIR/scripts/prepare_existing_deck.py" user_model.k --outdir template_build/<模板名称> --objective "制作 <模板用途> 模板", 将源文件和 include 闭包置于只读source/,只在独立working/(候选模板)与validation/目录工作。 该模式必须先完成结构化分析和模板适用性判断;会影响模板物理、参数接口或验收结论的问题,整个请求 最多集中询问 5 个关键问题。问题未回答前不得抽象最终模板或进行最终 L2 验证;用户确认后才写入template_spec.md、NOTES.md和模板文件。来源 deck 的验证结果不可直接替代模板候选的重新验证。 - 已有 deck 模式:用户提供主
.k/.key/.dyn并要求修改、调试、运行或修复。跳过 §1 需求头脑风暴、spec.md确认、模板/基准/学术门和新建流程,执行references/existing-deck-repair.md;仍须做 环境检查、knowledge/errors.md预检、L0/L1/L2、§5 交付报告和经验追加。无spec.md时保持 原 deck 的物理意图,仅解决用户目标范围的问题。先运行python "SKILL_DIR/scripts/prepare_existing_deck.py" user_model.k --objective "用户目标",不得直接修改原文件。 - 新建 deck 模式:无可用主 deck 或要求重新建模,执行 §2-§5。
会话首次运行:
python "SKILL_DIR/scripts/kagent_config.py"
默认求解器为 C:\Program Files\ANSYS Inc\v242\ansys\bin\winx64(LS-DYNA R14.1.1);用
LSDYNA_BIN、LSDYNA_MPIEXEC 或 ~/.lsdyna-kagent.json 覆盖。不得使用 R14 之后才支持的关键字或字段。
凡标注 R15 及以上的功能一律回避。手册缺失时运行 python "SKILL_DIR/scripts/fetch_manuals.py"。
2. 新建 deck:任务书与基准门
读取 references/brainstorm-protocol.md。它是 12 项收敛清单、6 类工况问题库、spec.md 模板、
模板适用性和文献/官方基准门的唯一细则。
- 缺 7 项关键输入(工况、目的/输出、几何/尺寸、材料、边界、初始条件、单位制)时,最多 5 轮、
每轮最多 4 个问题,按工况→物理→数值提问;给 2-4 个推荐选项。用户说“你定”时采用推荐值并标
[默认]。用户未指定单位制时默认mm-ton-s并声明;每轮做量级检查和收敛面板。 - 在选模板、生成网格、写任何卡前读取
knowledge/errors.md,将命中项写为预防检查;无命中也记录。 L0 前逐项核对;仅在新症状或预检未覆盖时重查。 - 完成模板适用性判断和基准方案。模板目录存在不等于模板可用:只有物理机制、建模方法、主要载荷/
边界、关注输出均能支撑当前骨架时可用;否则是无适用模板。
无适用模板时,必须先完成 LS-DYNA 官方资料/官方算例检索和学术文献检索,
不得用普通网页搜索、既有经验或相邻模板直接替代该双链路;
触发学术检索时必须有
research/literature-evidence.json和任务书文献摘要。爆炸焊接、界面波、飞板碰撞、爆轰耦合、 热-塑性-失效耦合且无专用模板时默认无适用模板。 - 模板速查:跌落
templates/drop/(mm-ton-s)、碰撞/压溃templates/crash/(mm-ton-s)、侵彻templates/penetration/(cm-g-us)、成形templates/forming/(mm-ton-s)、ALEtemplates/ale/(m-kg-s)、SPHtemplates/sph/(mm-kg-ms)。无相似案例时标no-similar-case,按任务书的M0~M3阶梯:M0 最小模型→M1 材料→M2 相互作用→M3 完整模型逐步验证。M0 先过 L0/L1 并核对单位、时间步和运动 方向;M1 用单单元/简化试件核对本构、失效/EOS、趋势或标定点;M2 检查穿透、初始间隙、反力、 能量传递和约束;M3 每次只增加一类复杂度并重跑 L0/L1,全部集成后才做 L2。关键无证据参数需给出 范围并做敏感性分析,报告参数区间、响应区间和结论稳定性。 - 将完整任务书写为工作目录
spec.md,状态“待用户确认”,展示内容和路径并问“按此任务书执行?”。 在收到用户明确同意前停止:不得生成、复制、修改或运行任何最终.k文件、网格 include 或写卡脚本, 也不得进入资料补全或生成阶段。 修改请求只更新spec.md后重新确认;同意后标“已确认(日期)”。 这是全流程唯一一次强制确认。无人值守才可不阻塞,所有缺项标[假设]并在最终报告显著提示。
What ships with it
60 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- knowledge/errors.md 7.7 KB
- knowledge/materials.json 8.8 KB
- references/brainstorm-protocol.md 16 KB
- references/existing-deck-repair.md 4.3 KB
- references/kfile-format.md 9.0 KB
- references/literature-parameter-routing.md 7.3 KB
- references/quality-gates.md 2.1 KB
- references/template-authoring-protocol.md 9.0 KB
- scripts/check_kfile.py 20 KB runs code
- scripts/fetch_manuals.py 1.3 KB runs code
- scripts/gen_mesh.py 17 KB runs code
- scripts/kagent_config.py 1.7 KB runs code
- scripts/manual_index.py 4.0 KB runs code
- scripts/parse_results.py 8.4 KB runs code
- scripts/prepare_existing_deck.py 5.4 KB runs code
- scripts/run_dyna.py 8.0 KB runs code
- scripts/setup_academic_mcp.py 11 KB runs code
- scripts/units.py 5.1 KB runs code
- templates/ale/ale.k 5.3 KB
- templates/ale/NOTES.md 4.7 KB
- templates/crash/crash.k 7.1 KB
- templates/crash/mesh_tube.k 203 KB
- templates/crash/NOTES.md 3.6 KB
- templates/drop/drop.k 4.8 KB
- templates/drop/mesh_block.k 8.5 KB
- templates/drop/mesh_plate.k 100 KB
- templates/drop/NOTES.md 3.9 KB
- templates/forming/build_deck.py 11 KB runs code
- templates/forming/forming.k 18 KB
- templates/forming/mesh_punch.k 64 KB
- templates/forming/mesh_strip.k 142 KB
- templates/forming/mesh_sup_left.k 48 KB
- templates/forming/mesh_sup_right.k 48 KB
- templates/forming/NOTES.md 5.1 KB
- templates/penetration/mesh_projectile.k 38 KB
- templates/penetration/mesh_target.k 374 KB
- templates/penetration/NOTES.md 4.2 KB
- templates/penetration/penetration.k 6.4 KB
- templates/sph/mesh_plate.k 100 KB
- templates/sph/mesh_sph.k 263 KB
- templates/sph/NOTES.md 3.8 KB
- templates/sph/sph.k 5.4 KB
- tests/test_academic_search_surface.py 3.0 KB runs code
- tests/test_existing_deck_repair.py 3.6 KB runs code
- tests/test_protocol_no_template_routing.py 4.2 KB runs code
- tests/test_setup_academic_mcp.py 12 KB runs code
- tests/test_skill_workflow_contract.py 6.9 KB runs code
- vendor/academic-search-mcp/academic_search_server.py 5.8 KB runs code
- vendor/academic-search-mcp/config.toml 249 B
- vendor/academic-search-mcp/README.md 953 B
- vendor/academic-search-mcp/requirements.txt 109 B
- vendor/academic-search-mcp/sources/__init__.py 352 B runs code
- vendor/academic-search-mcp/sources/arxiv.py 10 KB runs code
- vendor/academic-search-mcp/sources/crossref.py 4.3 KB runs code
- vendor/academic-search-mcp/sources/elsevier_common.py 1.5 KB runs code
- vendor/academic-search-mcp/sources/pubmed.py 7.8 KB runs code
- vendor/academic-search-mcp/sources/sciencedirect.py 4.8 KB runs code
- vendor/academic-search-mcp/sources/scopus.py 3.8 KB runs code
- vendor/academic-search-mcp/utils/__init__.py 359 B runs code
- vendor/academic-search-mcp/utils/config.py 1.4 KB runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 7d ago First seen · 139 lines · 106 tokens per session scan A 0a5a2c94307f
lsdyna-kfile is a skill published in the GitHub repository LLK-LL/K-agent (10 stars, last pushed 1mo ago), licensed MIT. It adds 106 tokens to every session and 3,211 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
patsnap-pharma-intelligence
Patsnap Pharma Intelligence MCP for AI agents. Search clinical trials, drugs, patents, papers, deals, FDA labels and more via 28 specialized tools.
patsnap-biologics-sequence
Patsnap Biologics & Sequence MCP for AI agents. Intelligent platform for biological sequences and macromolecules, covering antibody-antigen interaction discovery, full-lifecycle sequence search, sequence fetching and alignment, and modification status analysis.
patsnap-biological-modality
Biological sequence and modality intelligence via Patsnap MCP.
patsnap-chemical-molecular
Patsnap Chemical Molecular MCP for AI agents. Search 160M+ chemical structures, synthetic routes, and bioactivity data via specialized chemistry tools.
patsnap-scientific-translational-evidence
Patsnap Scientific & Translational Evidence MCP for AI agents. Retrieval platform focusing on scientific literature and translational outcomes, covering academic publication queries and translational medicine record tracking.
patsnap-target-disease
Patsnap Target & Disease MCP for AI agents. Target and disease profiling tool, covering target characterization, disease profiling, and epidemiology evidence retrieval.