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 Nero1688/claude-academic-skills --skill research-framework-figuregit clone --depth 1 https://github.com/Nero1688/claude-academic-skillsWrote 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/nero1688/claude-academic-skills/research-framework-figure)<a href="https://agentmods.dev/skills/nero1688/claude-academic-skills/research-framework-figure"><img src="https://agentmods.dev/badge/skills/nero1688/claude-academic-skills/research-framework-figure/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/nero1688/claude-academic-skills/research-framework-figure"><img src="https://agentmods.dev/badge/skills/nero1688/claude-academic-skills/research-framework-figure.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00546 | $0.02537 |
| Opus 5 | $0.00273 | $0.01269 |
| Sonnet 5 | $0.00109 | $0.00507 |
| Haiku 4.5 | $0.00055 | $0.00254 |
Grade A, and why
research-framework-figure 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 6d 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 — 126 lines — stays where its author put it; the contents beside it link to each section on GitHub.
研究架構圖產生器（Research Framework Figure）
與同族 skill 分工
| 需求 | 該用 |
|---|---|
| 已經有迴歸係數／CI，要畫統計結果圖（森林圖、交互作用圖、倒U轉折點、邊際效果） | management-figure |
| 還沒跑資料，要畫假說關係的概念圖（X→M→Y、H1/H2/H3） | 本 skill |
| 要決定用什麼識別策略（DiD／IV／RDD）、模型該怎麼設定 | causal-inference-architect |
| 要把圖放進簡報、決定簡報整體結構 | academic-pptx（本 skill 產圖供其嵌入） |
一句話：有係數了找 management-figure，還在畫假說找我。
Claude Code 環境呼叫同族技能須加 anthropic-skills: 前綴。
核心原則
- 模型邏輯由研究者定，我只負責畫。 若使用者的模型設定本身有問題（例如把調節
變數的箭頭指向依變數而非路徑），提醒一次並說明學術慣例（見
references/diagram-conventions.md），但最終照使用者決定畫。 - 字型誠實：SVG 的 font-family 一律寫成
Times New Roman, DFKai-SB, serif—— 西文在前、中文 fallback 到標楷體，這是 SVG 實現「中文標楷體＋英數 Times New Roman」 的正確作法。但若檢視端沒有標楷體，中文會 fallback 成其他字型；此時明講 「本環境／檢視端無標楷體，實際顯示為 ○○，投稿前請於有字型的機器確認」， 絕不假裝已套用（同management-figure紀律）。 - 產出必須可再編輯：一律給向量格式。SVG 給編修與插入 Word；PPTX 給口試前微調。 不以 PNG 為主要交付（點陣圖改不動，口試前臨時要改字會卡死）。
- 不塞爆版面：控制變數多於 12 項時，建議改列在圖下方註解或表格，不要硬塞進框內 （框會長到破壞版面比例）。這是版面判斷，要主動提醒。
- 產圖後必須實際檢視：產出 SVG 後，用瀏覽器或
--check幾何驗證確認無重疊、 無出界、箭頭端點正確，再交付。不要只憑「腳本沒報錯」就宣稱圖是好的。
工作流程
Step 1｜辨識模型類型
從使用者的假說敘述判斷屬於哪一種，對應到 template：
| 使用者的講法 | template |
|---|---|
| 「X 透過 M 影響 Y」「中介效果」，且每個依變數各有一組控制變數 | mediation_dual_controls（台灣商管論文最常見） |
| 「X 透過 M 影響 Y」，控制變數只列一組或不畫 | mediation |
| 「W 會調節 X 對 Y 的影響」「交互作用」 | moderation |
| 「中介效果會因 W 而異」「第一階段被調節」 | moderated_mediation |
| 「X→M1→M2→Y」「連續兩個中介」 | serial_mediation |
不確定時問一個問題：「你的控制變數是每條迴歸式各一組，還是全模型共用一組？」 ——這個答案直接決定要不要用雙控制框版式。
Step 2｜蒐集規格內容
需要問齊：自變數名稱、中介／調節變數（可多項）、應變數名稱、假說編號（H1/H2/H3…）、 控制變數清單（分別對應哪個依變數）。使用者若已有舊圖，照著複述一遍確認再畫。
Step 3｜寫 JSON 規格並產圖
規格格式見 references/framework-figure-spec.md。指令：
python scripts/framework_figure.py spec.json -o fig.svg
python scripts/framework_figure.py spec.json -o fig.pptx --format pptx
Step 4｜檢視與微調
產出後實際看圖（瀏覽器開 SVG，或把 SVG 包進 HTML 檢視）。檢查：
- 方框有無重疊、文字有無溢出框外
- 箭頭是否連到正確的框、假說標籤有沒有被線壓到
- 控制變數框是否過大而搶走視覺重心
不滿意就調
style區塊（字級、行高、邊距）或縮減條列文字後重出。
What ships with it
10 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.
- ATTRIBUTION.md 3.9 KB
- references/diagram-conventions.md 4.1 KB
- references/framework-figure-spec.md 5.5 KB
- scripts/examples/example-mediation-dual-controls.json 879 B
- scripts/examples/example-moderation.json 258 B
- scripts/examples/example-serial-mediation.json 373 B
- scripts/examples/README.md 1.2 KB
- scripts/framework_figure.py 23 KB runs code
- scripts/README.md 2.4 KB
- scripts/svg2drawingml.py 14 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.
- 6d ago First seen · 126 lines · 546 tokens per session scan A 98638ca88ee2
research-framework-figure is a skill published in the GitHub repository Nero1688/claude-academic-skills (6 stars, last pushed 8d ago), licensed MIT. It adds 546 tokens to every session and 2,537 once invoked, about $0.0027 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 2026-09-05.
Other skills, from other repositories
information-architecture
Design the structure of a website or product including sitemap, navigation, URL structure, content types, taxonomy, and labeling. Use this skill whenever the user asks to plan a sitemap, design navigation, structure URLs, define content types, build taxonomies, design site search, or organize content at the system…
journey-mapping
Build customer journey maps and service blueprints that visualize the end-to-end user experience including touchpoints, emotions, friction, and underlying systems. Use this skill whenever the user wants to map a customer journey, build a service blueprint, identify friction across an experience, align teams on the…
brand-style-guide
Build or audit a comprehensive brand style guide that documents the full brand system including story, logo system, color, typography, imagery, voice, applications, and dos/don'ts. Use this skill whenever the user wants to create brand guidelines, document an existing brand, build a brand book, audit an existing style…
visual-qa
Runs rigorous visual QA across web, terminal, and paginated surfaces with screenshot evidence and a verdict. Use for any UI build or change, or when asked whether a page, component, or TUI looks right.
frontend
Builds, styles, and polishes web UI and UX. Use for any frontend, page, component, styling, layout, animation, or visual-quality task, or when asked to make an interface look or feel a certain way.
interaction-skill
Layer A interaction-mechanics reference anchored to the beui.dev catalog. Stacks on any style skill whenever work adds or changes motion or interaction — micro-interactions, animated components, transitions, gestures, hover/press/state feedback, loading/success/error morphs, 'make it feel alive'. Mandates reading the…