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 kopp0510/claude-dd --skill tech-diagram-gifgit clone --depth 1 https://github.com/kopp0510/claude-ddWrote 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/kopp0510/claude-dd/tech-diagram-gif)<a href="https://agentmods.dev/skills/kopp0510/claude-dd/tech-diagram-gif"><img src="https://agentmods.dev/badge/skills/kopp0510/claude-dd/tech-diagram-gif/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/kopp0510/claude-dd/tech-diagram-gif"><img src="https://agentmods.dev/badge/skills/kopp0510/claude-dd/tech-diagram-gif.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00089 | $0.05128 |
| Opus 5 | $0.00044 | $0.02564 |
| Sonnet 5 | $0.00018 | $0.01026 |
| Haiku 4.5 | $0.00009 | $0.00513 |
Grade A, and why
tech-diagram-gif 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 3d 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 — 212 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Tech Diagram GIF — 技術圖表繪製與 GIF 匯出
不引入 Python/cairosvg 渲染管線的技術圖表流程:手寫 SVG(依 vendored 風格規範)→ 瀏覽器渲染自檢 → 匯出無縫循環 GIF。
渲染與匯出全程靠瀏覽器與 ffmpeg,不裝 cairosvg / Pillow 等套件;scripts/verify-geometry.py
是唯一的 Python,純標準庫、無 pip 依賴,只做座標算術,缺 python3 時退化為人工計算,不阻擋出圖。
風格規範 vendored 自 fireworks-tech-graph(MIT,見 LICENSE.txt);
只收編其 markdown 規範,不引入其 Python/cairosvg/FFmpeg 管線。
硬閘門
- 先判斷該不該畫 — 動筆前問一句:讀者從這張圖學到的,會比從一段文字或一張表格多嗎? 不會就別畫,直接寫清楚。純清單、簡單前後對照、單一形狀的「圖」一律退回文字。 使用者明確指定要圖時跳過此閘門,但仍可提一句更省的做法(提一次,不反覆)。
- 結構未確認前,不寫任何 SVG — 先用純文字把節點/分組/連線定案(工作流程第 1 步)。
- 交付物只有 GIF(使用者硬規則)。SVG 僅為中間產物:GIF 產出並驗證後即刪除, 不留在交付目錄。唯一例外:環境無 ffmpeg 時依第 5 步退化交付 SVG 並附裝法與資安說明 (SVG 格式可含 script,收件方有信任成本;本 skill 產的 SVG 零 script,交付前必驗證)。
- 渲染自檢不可省 — 沒親眼看過渲染結果不得宣稱完成。
工作流程
1. 結構確認(純文字,不畫圖)
把要畫的內容整理成清單給使用者確認,格式:
節點(N):① 名稱(語意桶)→ ② …
分組(M):分組名 / 涵蓋節點
連線與語意:A→B 主流程(金)/ C⇢D 條件檢查(虛線)/ 自迴圈…
清單交出去前先過下方 Taste Gate 的「結構清單」組 —— 數量上限在這一步就數得出來 (節點/連線/分組三個數字見該組,本節不複製),此時超標只要改清單; 等到第 4 步才發現,只能整份重來。 使用者確認或修改 → 定案才進下一步。內容事實(節點名稱、流程順序)不確定時回問,不自行腦補。
2. 選風格與動畫模式
| 風格 | 檔案 | 何時用 |
|---|---|---|
| Style 8 Dark Luxury(預設) | references/style-8-dark-luxury.md |
架構、流程、資料流、拓撲 — 通用首選 |
| Style 2 Dark Terminal | references/style-2-dark-terminal.md |
開發者風格;UML 類(ER/時序/狀態機/use case)優先用此 — Style 8 深底吃 UML 細節與人形,比較表也不適合 |
| Style 11 Event Transit | references/style-11-event-transit.md |
事件流/串流拓撲(Kafka/Pulsar/NATS、消費者群、DLQ)— 提到「事件流」「地鐵圖」「Kafka 拓撲」時用 |
| Style 12 Ops Pulse | references/style-12-ops-pulse.md |
事故排查/SLO review(golden signals、critical path、trace 瀑布)— 提到「事故圖」「golden signals」「reliability」時用 |
動畫兩種模式(詳見 references/motion-narrative.md):
- 循環流動(預設):小球沿線跑(
animateMotion)或 dash 脈衝 - 建置 → 營運(敘事):開場無連線 → 逐條畫入 → 營運流動 → 淡出循環 — 對外展示/README hero 用
使用者無明示偏好時用預設,不必多問;圖型落在特定風格的甜蜜區時主動建議換風格(一句話,不反覆)。
3. 手寫 SVG
- 版面規則照
references/svg-layout-best-practices.md(間距 ≥80px、正交轉角、標籤偏移、z-order; 該檔僅取版面規則,其 cairosvg/PNG 匯出段落不適用本 skill) - 節點複雜或含 DB/佇列/使用者/決策時,用
references/icons.md的語意圖形 (DB 圓柱、佇列管、LLM 雙框、Agent 六角形、User 人形、決策菱形、文件摺角)取代一律圓角矩形 - 量化預算三組 —— showcase profile 表(交叉/折數/繞路比/間距)、「元素數量預算」、
「連線可量測規則」—— 數字全在
references/composition-quality-contract.md,本節不複製。 撰寫時開那張表照著寫,操作用的逐項版本見下方 Taste Gate;兩處不符一律以 contract 為準。 超出 → 調節點位置,或拆成總覽 + 細節兩張;不縮字級、不壓間距硬塞 - 節點間距 80px vs 40px、標籤三個間隙數字各量什麼 —— 裁決與說明見 contract 「與上表衝突時取嚴」節,不要自行挑一個用
- 節點、容器、連線、標籤遮罩一律加
data-role(node/container/edge/mask) ——scripts/verify-geometry.py優先讀這個標記;沒有標記時它會退化用畫法猜並印警告, 猜錯就是靜默漏檢(實測踩過:矩形節點被當成菱形而誤報溢出;漸層畫布讓兩項遮罩檢查 一起空轉卻印「全部通過」) - 版面順序與走廊:先排容器與列才排線、保留跨層走廊、legend 不進流程走廊
- 色票/節點語意色桶照所選風格檔;畫布建議
viewBox 0 0 1440 1080— 注意風格檔的字級/間距以 960 寬為基準,用 1440 畫布時需等比放大(約 ×1.5) - 字體堆疊必含跨平台 CJK 後備(以本段為準,覆蓋風格檔的 PingFang SC):
襯線
Georgia,'Times New Roman','Songti TC','Noto Serif CJK TC',serif、 無襯線-apple-system,'Helvetica Neue','PingFang TC','Noto Sans CJK TC',sans-serif(macOS 已驗證;Linux 走 Noto 後備,未實測) - 小球規則:核心 r4 + 光暈 r8 opacity 0.22、顏色跟隨箭頭語意、等速(dur ∝ 路徑長)、 長路徑放 2 顆錯開半週期
- 所有動畫 dur 與 begin 必須整除同一個總循環長(如 7.2s)— GIF 才能無縫循環
- dash 脈衝模式:實線疊亮色
stroke-dasharray: 7 41動stroke-dashoffset;位移量須為 dasharray 週期整數倍
What ships with it
13 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.
- LICENSE.txt 1.9 KB
- references/composition-quality-contract.md 7.2 KB
- references/icons.md 15 KB
- references/motion-narrative.md 2.6 KB
- references/style-11-event-transit.md 3.6 KB
- references/style-12-ops-pulse.md 3.3 KB
- references/style-2-dark-terminal.md 2.9 KB
- references/style-8-dark-luxury.md 7.6 KB
- references/svg-layout-best-practices.md 4.4 KB
- scripts/CLAUDE.md 5.0 KB
- scripts/fixtures/sample-flow.svg 12 KB
- scripts/test-verify-geometry.py 13 KB runs code
- scripts/verify-geometry.py 35 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.
- 3d ago Changed · +84 lines 7f1e2a72d4d6
- 11d ago First seen · 128 lines · 89 tokens per session scan A 8a705d560c2f
tech-diagram-gif is a skill published in the GitHub repository kopp0510/claude-dd (6 stars, last pushed 3d ago), licensed MIT. It adds 89 tokens to every session and 5,128 once invoked, about $0.0004 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
frontend-code-review
Trigger when the user requests a review of frontend files (e.g., .tsx, .ts, .js). Support both pending-change reviews and focused file reviews while applying the checklist rules.
design-patterns
Agents should invoke this skill when choosing patterns, designing traits/interfaces/components, deciding abstraction boundaries, evaluating dependency injection/callbacks, or comparing implementation approaches in Rust, TypeScript/React, or Django/Python.
ln-75-architecture-diagram-builder
Creates current or target architecture diagrams as the primary deliverable. Not for UI design, architecture audits, or invented structure.
code-architect
Design a feature architecture by analyzing existing codebase patterns and conventions, then provide a comprehensive implementation blueprint with specific files to create or modify, component designs, data flows, and a build sequence. Use this skill when the user asks for an architecture design, an implementation plan…
spotpatch
A development workflow for React that connects a selected page element to its JSX or TSX source code and gathers context for review.
docs-release-notes
Use when a change needs a user-visible release-note, changelog, or changeset entry — "add a release note", "add a changeset for this", "what goes in the changelog?", "write up what shipped", "note this for the next release" — or when finalizing a branch whose customer-visible features, bug fixes, or UI changes should…