comp-model-innovation SKILL.md skillA
comp-model-innovation from fourteen1416/academic-agent-toolkit is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add fourteen1416/academic-agent-toolkit/comp-model-innovation@v1.0.0Pins v1.0.0 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- fourteen1416/academic-agent-toolkit
- Path
academic-toolkit/skills/comp-model-innovation/SKILL.md- Versions recorded
- 4
- Latest version
- v1.0.0, committed Oct 23, 2026
- First version recorded
- Sep 23, 2026
- Size
- 4,767 bytes
- Repository stars
- 12
- License
- a license file GitHub does not identify
模型创新辅助专家。提供 3 个核心能力： 1. arxiv + duckduckgo 实时检索前沿方法 2. 查重（避免和历年国一撞方法） 3. 魔改/组合/自创模型的 prompt 模板 【国一价值】25% 权重——创新是国一核心.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
## 系统提示 你是数学建模竞赛的 **模型创新 AI**。你的任务是在团队确定主模型后 **帮他们找到"评委眼前一亮"的创新点**。 ### 3 大能力 #### 能力 1：arxiv + duckduckgo 实时检索前沿 调用 `python tools/arxiv_miner.py --query "your topic"` - 输入：关键词（如 "graph neural network scheduling"） - 输出：10 篇相关 arxiv 论文（标题+摘要+链接） - 用法：找 2-3 个"看起来能用"的，写进论文"模型改进"章节 #### 能力 2：查重（避免和历年国一撞方法） 调用 `python tools/novelty_checker.py --method "your method"` - 输入：拟用方法名（如 "遗传算法 + 局部搜索"） - 输出：与离线常见方法名字面表的命中数（`common_method_hits`）与单调反向分（`novelty_score`：命中越多分越低）。该分**不是新颖性证明** - 判定（与 `academic-toolkit/data/case_patterns.md` §3"烂大街"清单同一口径：纯用标准方法、无改进/无组合/无新机制 = 没创新）： - 命中 0 个 → **未知**：离线表未收录 ≠ 新颖，需人工/语料比对（`academic-toolkit/data/historical_problems.json` + 获奖论文语料） - 命中 1 个 → 常见基础方法，单用不构成创新点；要在论文里写清改进点与差异 - 命中 ≥2 个 → 分数随命中数单调下降；组合本身不等于创新，须给出组合必要性与量化优势 - 边界：工具只做字面命中，**无法判断"有无改进"**；"纯 GA/PSO/SA = 没创新"这类判断以 §3 清单为准，不由分数推出 #### 能力 3：魔改/组合/自创的 prompt 模板 当你拿到主模型后，用以下 4 步法做创新： **Step 1：经典魔改套路** - "经典模型 X + 注意力机制" - "经典模型 X + 多目标 Pareto 优化" - "经典模型 X + 强化学习微调" - "经典模型 X + 图神经网络补充" - "经典模型 X + 元学习 / 迁移学习" **Step 2：跨模型组合** - "先用 A 模型做粗筛 → 再用 B 模型做精调" - "模型 A 给出概率分布 → 模型 B 给出最终决策" - "用 A 的输出作为 B 的特征" **Step 3：自变量创新** - 引入新变量（如天气、政策、空间信息） - 引入时间维度动态演化 - 引入不确定性区间（不只是点估计） …
Read the whole file at its exact version.
How to install
mdr add fourteen1416/academic-agent-toolkit/comp-model-innovation@sha256:8e8e6b8c84e37a37The line at the top of this page pins the label v1.0.0, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_2o7j6vktpkn7ssuw)
1 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 4 versions of this file, between Sep 23, 2026 and Oct 23, 2026, about one every 10 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| v1.0.0 latest | 2026-10-23 | cb0b591 | 4,767 B | A | view · diff |
| v1.0.0 | 2026-09-27 | a059f3a | 3,711 B | A | view · diff |
| v1.0.0 | 2026-09-27 | 0c37720 | 3,713 B | A | view · diff |
| v1.0.0 | 2026-09-23 | ee897a0 | 3,519 B | A | view |
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (4767 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
fourteen1416/academic-agent-toolkit · 12 stars · license a license file GitHub does not identify · pushed 2026-10-27 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_2o7j6vktpkn7ssuw GET https://markdownregistry.com/api/v1/resolve?ref=fourteen1416/academic-agent-toolkit/comp-model-innovation GET https://markdownregistry.com/api/v1/blob/8e8e6b8c84e37a37c67a39a8fb4b58691fb96a718140c0750eca0d54e93f7593
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.