create-glyph
About
This skill creates R-based pictogram glyphs for entity icons in the visualization layer using ggplot2 and a primitives library. It handles the entire workflow from concept and color strategy to registration, pipeline rendering, and verification of the neon-glow output. Use it when adding a new entity needing a visual icon, replacing an existing glyph, or batch-creating glyphs for a new domain.
Quick Install
Claude Code
Recommendednpx skills add pjt222/agent-almanac -a claude-code/plugin add https://github.com/pjt222/agent-almanacgit clone https://github.com/pjt222/agent-almanac.git ~/.claude/skills/create-glyphCopy and paste this command in Claude Code to install this skill
Documentation
造符
於 viz/ 視層造 R 基象圖符為技、代、團標。各符為純 ggplot2 函,於 100x100 畫繪可識形、以霓暈效渲為透 WebP。
用
- 新技、代、團已加須視標
- 存符須換或重設
- 為新技域批造符
- 試視喻為項概
入
- 必:項型 —
skill、agent、team - 必:項 ID(如
create-glyph、mystic、r-package-review)與域(技時) - 必:視概——符當繪何
- 可:考符以評複
- 可:自定
--glow-sigma值(默:4)
行
一:概——設視喻
識所符項並擇視喻。
- 讀項源以曉核概:
- 技:
skills/<id>/SKILL.md - 代:
agents/<id>.md - 團:
teams/<id>.md
- 技:
- 擇喻型:
- 實物:實驗用瓶、安用盾
- 抽符:併用箭、迭用螺
- 合:合 2-3 簡形(如文+筆)
- 參存符以校複:
Complexity Tiers:
+----------+--------+-------------------------------------------+
| Tier | Layers | Examples |
+----------+--------+-------------------------------------------+
| Simple | 2 | glyph_flame, glyph_heartbeat |
| Moderate | 3-5 | glyph_document, glyph_experiment_flask |
| Complex | 6+ | glyph_ship_wheel, glyph_bridge_cpp |
+----------+--------+-------------------------------------------+
- 定函名:
glyph_<descriptive_name>(snake_case、獨)
得: 形之明心圖含 2-6 計層。
敗: 概過抽→退至相關實物。參同域存符為靈。
二:組——書符函
書生 ggplot2 層之 R 函。
-
函簽(不變約):
glyph_<name> <- function(cx, cy, s, col, bright) { # cx, cy = center coordinates (50, 50 on 100x100 canvas) # s = scale factor (1.0 = fill ~70% of canvas) # col = domain color hex (e.g., "#ff88dd" for design) # bright = brightened variant of col (auto-computed by renderer) # Returns: list() of ggplot2 layers } -
施比因
* s於諸寸為一致:r <- 20 * s # radius hw <- 15 * s # half-width lw <- .lw(s) # line width (default base 2.5) lw_thin <- .lw(s, 1.2) # thinner line width -
以可用原建幾:
幾 用 ggplot2::geom_polygon(data, .aes(x, y), ...)填形 ggplot2::geom_path(data, .aes(x, y), ...)開線/曲 ggplot2::geom_segment(data, .aes(x, xend, y, yend), ...)線段、箭 ggplot2::geom_rect(data, .aes(xmin, xmax, ymin, ymax), ...)矩 ggforce::geom_circle(data, .aes(x0, y0, r), ...)圓 -
施色策:
Alpha Guide: +----------------------+------------+--------------------------+ | Purpose | Alpha | Example | +----------------------+------------+--------------------------+ | Large fill (body) | 0.08-0.15 | hex_with_alpha(col, 0.1) | | Medium fill (accent) | 0.15-0.25 | hex_with_alpha(col, 0.2) | | Small fill (detail) | 0.25-0.35 | hex_with_alpha(bright, 0.3) | | Outline stroke | 1.0 | color = bright | | Secondary stroke | 1.0 | color = col | | No fill | --- | fill = NA | +----------------------+------------+--------------------------+ -
返平
list()層(渲器迭而以暈包各) -
按項型置函於合原檔:
- 技:19 原檔按域組
- 代:
viz/R/agent_primitives.R - 團:
viz/R/team_primitives.R
得: 返 2-6 ggplot2 層之可用 R 函。
敗: ggforce::geom_circle 誤→保 ggforce 裝。坐偏→記畫 100x100、(0,0) 於左下。交互試:
source("viz/R/utils.R"); source("viz/R/primitives.R")
layers <- glyph_<name>(50, 50, 1.0, "#ff88dd", "#ffa8f0")
p <- ggplot2::ggplot() + ggplot2::coord_fixed(xlim=c(0,100), ylim=c(0,100)) +
ggplot2::theme_void()
for (l in layers) p <- p + l
print(p)
三:註——映項至符
於合符映檔加項符映。
技: 開 viz/R/glyphs.R、於域塊內按字母序加 "skill-id" = "glyph_function_name",
代: 開 viz/R/agent_glyphs.R、於 AGENT_GLYPHS 加同樣項
團: 開 viz/R/team_glyphs.R、於 TEAM_GLYPHS 加同樣項
驗目列無重 ID。
得: 合 *_GLYPHS 列含新映。
敗: 構報「No glyph mapped」→復察項 ID 合備與庫。
四:備——加標項
於合備檔註標。
技: viz/data/icon-manifest.json
{
"skillId": "skill-id",
"domain": "domain-name",
"prompt": "<domain basePrompt>, <descriptors>, dark background, vector art",
"seed": <next_seed>,
"path": "public/icons/cyberpunk/<domain>/<skill-id>.webp",
"status": "pending"
}
代: viz/data/agent-icon-manifest.json(用 agentId 與 public/icons/cyberpunk/agents/<agent-id>.webp)
團: viz/data/team-icon-manifest.json(用 teamId 與 public/icons/cyberpunk/teams/<team-id>.webp)
得: 有效 JSON 含新項置於同型兄弟。
敗: 驗 JSON 文法。常誤:末陣元後遺逗、缺引。
五:渲——生標
行標流渲新符。恆用 build.sh 為入——理平台辨與 R 二擇。見 render-icon-pipeline 全旗考與流架。
# From project root — renders all palettes, standard + HD, skips existing icons
bash viz/build.sh --only <domain> --skip-existing # skills
bash viz/build.sh --type agent --only <id> --skip-existing # agents
bash viz/build.sh --type team --only <id> --skip-existing # teams
# Dry run first:
bash viz/build.sh --only <domain> --dry-run
build.sh 行全流(palette → data → manifest → render → terminal glyphs)。
出位:
- 技:
viz/public/icons/<palette>/<domain>/<skill-id>.webp - 代:
viz/public/icons/<palette>/agents/<agent-id>.webp - 團:
viz/public/icons/<palette>/teams/<team-id>.webp
得: 誌顯 OK: <entity> (seed=XXXXX, XX.XKB) 而 WebP 檔存。
敗:
"No glyph mapped"— 步三映缺或錯"Unknown domain"— 域不於palettes.R之get_palette_colors()- R 包誤 — 先行
install.packages(c("ggplot2", "ggforce", "ggfx", "ragg", "magick")) - 渲崩→交互試符函
六:驗——視察
察渲出合質準。
-
驗檔存且寸合:
ls -la viz/public/icons/cyberpunk/<type-path>/<entity-id>.webp # Expected: 15-80 KB typical range -
以像視開 WebP 察:
- 形於全寸(1024x1024)清讀
- 霓暈存而不蓋
- 背透(無黑白矩)
- 畫邊無裁
-
於小寸察(力圖中標渲於~40-160px):
- 形仍可識
- 細不化噪
- 暈不蓋形
得: 清、可識之象圖含透背之均霓暈。
敗:
- 暈強→以
--glow-sigma 2重渲 - 暈弱→以
--glow-sigma 8重渲 - 小寸不讀→簡符、粗筆、增
.lw(s, base) - 邊裁→減形寸或移中
七:迭——若需則精
調而重渲。
-
常調:
- 粗筆:增
.lw(s, base)— 試base = 3.0或3.5 - 更顯填:alpha 自 0.10 增至 0.15-0.20
- 形比:調
s之乘(如20 * s->24 * s) - 加/減細層:總層保於 2-6
- 粗筆:增
-
改後重渲:刪存標、用合構令重渲
-
滿時驗備態顯
"done"
得: 終標通步六諸察。
敗: 3+ 迭後符仍不讀→考全異視喻(返步一)。
考
域與項色盤
58 域色(技)定於 viz/R/palettes.R(唯真源)。代與團色亦理於 palettes.R。cyberpunk 色盤於 get_cyberpunk_colors()。
source("viz/R/palettes.R")
get_palette_colors("cyberpunk")$domains[["design"]]
get_palette_colors("cyberpunk")$agents[["mystic"]]
get_palette_colors("cyberpunk")$teams[["tending"]]
加新域時於 palettes.R 三處加:PALETTE_DOMAIN_ORDER(字母)、get_cyberpunk_colors() 域列、行 bash viz/build.sh 重生。
符函錄
- 技:
viz/R/primitives.R至viz/R/primitives_19.R(按域組) - 代:
viz/R/agent_primitives.R - 團:
viz/R/team_primitives.R
助函
| 函 | 簽 | 目 |
|---|---|---|
.lw(s, base) | (scale, base=2.5) | 比覺線寬 |
.aes(...) | ggplot2::aes 別 | 簡美映 |
hex_with_alpha(hex, alpha) | (string, 0-1) | 加 alpha 於 hex 色 |
brighten_hex(hex, factor) | (string, factor=1.3) | 亮 hex 色 |
dim_hex(hex, factor) | (string, factor=0.4) | 暗 hex 色 |
驗
- 符函循
glyph_<name>(cx, cy, s, col, bright) -> list()簽 - 諸寸用
* s比因 - 色策填用
col、廓用bright、透用hex_with_alpha() - 函置於合項型與域之原檔
- 符映項加於合
*_glyphs.R - 備項含正項 ID、路、
"status": "pending" - 構令無誤行(先試行)
- 渲 WebP 存於期路
- 檔寸於期範(15-80 KB)
- 標於 1024px 與~40px 顯寸皆清讀
- 透背(符後無實矩)
- 成渲後備態更為
"done"
忌
- 忘
* s:硬碼素值比變時破 - 畫原惑:(0,0) 於左下、非左上
- 雙暈:渲器已施
ggfx::with_outer_glow()、勿於符函內加暈 - 層過多:過 8 層渲慢且視噪
- ID 不合:符映、備、庫之項 ID 須皆合
- JSON 末逗:備為嚴 JSON、末陣元後無逗
- 域色缺:域不於
get_cyberpunk_colors()→渲誤 - 誤原檔:技於
primitives*.R、代於agent_primitives.R、團於team_primitives.R
參
- enhance-glyph — 精存符視質
- audit-icon-pipeline — 察缺符標
- render-icon-pipeline — 全端對端行渲流
- ornament-style-mono — 補 AI 基像生
- ornament-style-color — 施於符飾填決之色理
- create-skill — 觸符造之父流
GitHub Repository
Related Skills
content-collections
MetaThis skill provides a production-tested setup for Content Collections, a TypeScript-first tool that transforms Markdown/MDX files into type-safe data collections with Zod validation. Use it when building blogs, documentation sites, or content-heavy Vite + React applications to ensure type safety and automatic content validation. It covers everything from Vite plugin configuration and MDX compilation to deployment optimization and schema validation.
polymarket
MetaThis skill enables developers to build applications with the Polymarket prediction markets platform, including API integration for trading and market data. It also provides real-time data streaming via WebSocket to monitor live trades and market activity. Use it for implementing trading strategies or creating tools that process live market updates.
creating-opencode-plugins
MetaThis skill helps developers create OpenCode plugins that hook into 25+ event types like commands, files, and LSP operations. It provides the plugin structure, event API specifications, and implementation patterns for JavaScript/TypeScript modules. Use it when you need to intercept, monitor, or extend the OpenCode AI assistant's lifecycle with custom event-driven logic.
sglang
MetaSGLang is a high-performance LLM serving framework that specializes in fast, structured generation for JSON, regex, and agentic workflows using its RadixAttention prefix caching. It delivers significantly faster inference, especially for tasks with repeated prefixes, making it ideal for complex, structured outputs and multi-turn conversations. Choose SGLang over alternatives like vLLM when you need constrained decoding or are building applications with extensive prefix sharing.
