Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Get Started Free →

Xiaomacoltai/建模手 — measured AI agent skill | DecimalAI

xiaomacoltai/建模手

NewSkillScoreMaintained

数学建模的题目理解、模型选择和算法设计阶段。输出题目分析报告与术语表格。

scienceby xiaomacoltai

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

Research pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/xiaomacoltai-a7a45c/SKILL.md

decimalai skills pull xiaomacoltai-a7a45c --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/xiaomacoltai-a7a45c/SKILL.md https://app.decimal.ai/s/xiaomacoltai-a7a45c/SKILL.md

Raw file: /s/xiaomacoltai-a7a45c/SKILL.md ↗ · machine-readable summary at /s/xiaomacoltai-a7a45c.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 4h ago

—+15% tokens0% turns73% of whole cases pass with skill (16/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-03	✗→✓	▲ Improved	110%	0%
case-11	✗→✓	▲ Improved	-12%	0%
case-07	✗→✓	▲ Improved	12%	0%
case-09	✗→✓	▲ Improved	11%	0%
case-10	✗→✓	▲ Improved	-29%	0%

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v5 — 8/17/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

xiaomacoltai / math-modeling-skill

references/roles/%E5%BB%BA%E6%A8%A1%E6%89%8B/SKILL.md

View on GitHub ↗

Imported 8/17/2026Version v2Bundled 5 files

Skill contents

references/

SKILL.md

建模手

路径

ROLE_ROOT：本文件所在目录。
SKILL_ROOT：ROLE_ROOT/../../..，只读。
PROJECT_ROOT：用户项目目录，所有产物只写这里。

固定产物

PROJECT_ROOT/题目分析报告.md
PROJECT_ROOT/术语表格.md

只使用上述两个文件；合同内容作为分析报告的一部分。

执行顺序

读取题目与全部附件，建立问题、目标、约束、数据字段和输出要求清单。
数据驱动地检查缺失、异常、量纲、时间与空间范围。
在理解题目后写出每个子问题应回答的结论类型，再评估候选模型。
每个子问题最多选两个独立模型体系，说明模型、算法、公式、参数、输入输出、验证方式和风险；同一物理机制的不同近似或精度展开按一个模型族计数。
使用 ../../../tools/paper_search/SKILL.md 搜索理论依据；引用必须可追溯。
写入两个固定产物并按 references/质检清单.md 完成作者自检。
立即派发未参与编写的质检 Subagent 执行 M1 建模终检；未返回 PASS 不得进入编程阶段或宣称建模交付完成。

独立验收

M1 的输入包括题目与附件、题目分析报告.md、术语表格.md、文献来源和作者自检结果。质检 Subagent 只读核对子问题覆盖、公式与符号、单位与约束、假设依据、模型数量、可实现性、验证方案和引用可追溯性，按 ../../../references/Subagent调度.md 返回证据回执。失败时由建模手修正两个既有产物，再复验 M1。

何时加载

| 情形 | 读取 | |---|---| | 开始分析 | references/工作流程.md | | 写分析报告前确定合同 | references/前置合同.md | | 设计模型组合 | references/建模设计理论.md | | 需要常见问题模式 | references/常见模式.md | | 需要具体算法 | ../../../references/算法索引.md，再读取匹配的 ../../../assets/*.md | | 交付前 | references/质检清单.md | | 独立验收 | ../../../references/Subagent调度.md |

选择原则

模型必须覆盖题目约束、数据特征和评价目标。
避免直接套用常见简单模型；优先寻找与题目结构紧密相关、可验证且可实现的改进。
复杂度由问题需要决定，不把复杂度本身当作独立优点。
不因两个模型结论相似而机械删除模型。

若编程手反馈模型不可实现，基于具体报错和数据证据修订两个既有产物，不另建文件。

Want to see if xiaomacoltai/建模手 would activate on your phrasing? Try it before you install — no LLM call, no signup.

Other measured skills in the registry, with their headline benchmark lift.