О программе
Этот навык улучшает существующие пиктограммы на основе R, диагностируя и исправляя визуальные проблемы, такие как плохие пропорции, проблемы с читаемостью или несбалансированные эффекты свечения. Он проводит разработчиков через аудит глифа, внесение целевых изменений в функции и сравнение результатов "до" и "после". Используйте его, когда глифы плохо отображаются в малых размерах, имеют неясные визуальные метафоры или требуют обновления после изменений палитры или конвейера обработки.
Быстрая установка
Claude Code
Рекомендуетсяnpx 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/enhance-glyphСкопируйте и вставьте эту команду в Claude Code для установки этого навыка
Документация
Enhance Glyph
Improve existing pictogram glyph in viz/ viz layer — audit how it renders, diagnose issues, apply tight fixes, re-render, and compare before/after. Works for skill, agent, team glyphs.
When Use
- Glyph renders poor at small size (details lost, shapes merge)
- Glyph visual meaning unclear or not match entity
- Glyph has proportion issue (too big, too small, off-center)
- Neon glow eats glyph or too weak
- Glyph good in one palette, poor in others
- Batch fix after new palette or new render pipeline
Inputs
- Required: Entity type —
skill,agent, orteam - Required: Entity ID of glyph to enhance (e.g.,
commit-changes,mystic,tending) - Required: Specific issue to fix (readability, proportions, glow, palette compat)
- Optional: Reference glyph that shows target quality
- Optional: Target palette(s) to tune for (default: all palettes)
Steps
Step 1: Audit — Check Current State
Look at current glyph; spot specific issues.
- Find glyph function by entity type:
- Skills:
viz/R/primitives*.R(19 domain-grouped files), mapped inviz/R/glyphs.R - Agents:
viz/R/agent_primitives.R, mapped inviz/R/agent_glyphs.R - Teams:
viz/R/team_primitives.R, mapped inviz/R/team_glyphs.R
- Skills:
- Read glyph function to grasp shape:
- How many layers?
- What primitives call?
- What scale factors and place?
- View rendered output:
- Skills:
viz/public/icons/cyberpunk/<domain>/<skillId>.webp - Agents:
viz/public/icons/cyberpunk/agents/<agentId>.webp - Teams:
viz/public/icons/cyberpunk/teams/<teamId>.webp - If can, check 2-3 other palettes for cross-palette render
- View at icon size (~48px in graph) and panel size (~160px in detail panel)
- Skills:
- Score glyph on quality dimensions:
Glyph Quality Dimensions:
+----------------+------+-----------------------------------------------+
| Dimension | 1-5 | Assessment Criteria |
+----------------+------+-----------------------------------------------+
| Readability | | Recognizable at 48px? Clear at 160px? |
| Proportions | | Well-centered? Good use of the 100x100 canvas?|
| Metaphor | | Does the shape clearly represent the entity? |
| Glow balance | | Glow enhances without overwhelming? |
| Palette compat | | Looks good across cyberpunk + viridis palettes?|
| Complexity | | Appropriate layer count (not too busy/sparse)? |
+----------------+------+-----------------------------------------------+
- Spot 1-2 dimensions with lowest scores — these are fix targets
Got: Clear pick of what wrong with glyph and which dimension to fix. Audit specific: "proportions: glyph uses only 40% of canvas" not "looks bad."
If fail: Glyph function missing or entity not in its *_glyphs.R map? Glyph maybe not made yet — use create-glyph instead.
Step 2: Diagnose — Root Cause
Find why issues exist.
- For readability issues:
- Too many fine details that merge at small size?
- Weak contrast between glyph elements?
- Lines too thin (< 1.5
sizeat s=1.0)? - Elements too close?
- For proportion issues:
- Scale factor
stoo small or too big? - Center off from (50, 50)?
- Elements past safe area (10-90 range)?
- Scale factor
- For glow issues:
- Glyph stroke width mixes with
ggfx::with_outer_glow():- Thin lines: glow make fuzzy
- Thick fills: glow add bloom
- Many overlap elements: stacked glow make hot spots
- Glyph stroke width mixes with
- For palette compat issues:
- Glyph use hardcoded colors, not
col/brightparams? - Low-contrast palettes (cividis, mako) hide glyph?
- Glyph needs color variation some palettes miss?
- Glyph use hardcoded colors, not
- Write specific root cause for each issue
Got: Root causes that point to code change. "Glyph too small" -> "scale factor 0.6 but should be 0.8." "Glow eats" -> "three overlap filled polygons each make glow."
If fail: Root cause not clear from code read? Render glyph alone with different params to isolate. Use render_glyph() with one glyph to test.
Step 3: Modify — Apply Tight Fixes
Edit glyph function to fix diagnosed issues.
- Open file with glyph function
- Apply fixes matched to diagnosis:
- Scale/proportion: Tune
smultiplier or element offsets - Readability: Simplify complex parts, thicker stroke, add space
- Glow balance: Fewer overlap filled areas, use outlines where fills bloom
- Palette compat: All colors from
col/brightparams, add alpha for depth
- Scale/proportion: Tune
- Follow glyph function contract:
glyph_name <- function(cx, cy, s, col, bright) { # cx, cy = center (50, 50) # s = scale (1.0 = ~70% of canvas) # col = domain color, bright = brightened variant # Returns: list() of ggplot2 layers } - Keep function signature — do not change params
- Keep changes tight: fix diagnosed issues, do not redesign whole glyph
Got: Modified glyph function that fix specific issues from Steps 1-2. Changes tight and minimal — enhance, not redesign.
If fail: Fixes make other dimensions worse (e.g., fixing proportions breaks readability)? Revert and try other path. If glyph need full redesign, use create-glyph instead.
Step 4: Re-render — Make New Icons
Render modified glyph and check fix. Always use build.sh — it handles platform detect and R binary pick. See render-icon-pipeline for full flag list.
-
Re-render by entity type:
# From project root — use --no-cache to force re-render of modified glyph bash viz/build.sh --only <domain> --no-cache # skills bash viz/build.sh --type agent --only <id> --no-cache # agents bash viz/build.sh --type team --only <id> --no-cache # teams -
Check output files at expected path for each palette
-
Check file sizes — icons should be 2-15 KB (WebP):
- Under 2 KB: glyph maybe too simple or render fail
- Over 15 KB: glyph maybe too complex (too many layers)
Got: Fresh icon files made for all palettes. File sizes in expected range.
If fail: Build script errors? Check R console output for specific error. Common: missing close paren in glyph function, ref undefined primitives, or return non-list from function. If render OK but output blank, glyph layers maybe outside canvas bounds.
Step 5: Compare — Before/After Check
Check fix improved target dimensions.
- Compare old and new renderings:
- View cyberpunk palette at icon (48px) and panel (160px) sizes
- View at least 2 other palettes (one light like turbo, one dark like mako)
- Re-score quality dimensions from Step 1:
- Target dimensions should improve by at least 1 point
- Non-target dimensions should not drop
- If glyph used in force-graph, test there:
- Start HTTP server:
python3 -m http.server 8080fromviz/ - Load graph and find entity node
- Check icon renders right at default zoom and zoomed in
- Start HTTP server:
- Write what changes made and gain reached
Got: Measurable gain on target dimensions with no drop on others. Glyph looks better at both sizes and across palettes.
If fail: Gain tiny or drop happens? Revert changes and rethink diagnosis. Sometimes old glyph limits come from metaphor, not code — then metaphor itself may need change (escalate to create-glyph).
Validation Checklist
- Current glyph audited with specific issue call
- Root cause found for each issue
- Changes tight to diagnosed issues (not over-edited)
- Glyph function contract kept (signature unchanged)
- Icons re-rendered for all palettes
- Before/after compare shows gain on target dimensions
- No drop on non-target dimensions
- File sizes in expected range (2-15 KB WebP)
- Glyph renders right in force-graph context (if used)
Pitfalls
- Over-enhancement: Fix one issue then tweak everything else. Stick to diagnosed issues
- Break contract: Change function signature break render pipeline. 5-param contract is fixed
- Palette-specific tune: Make glyph perfect for cyberpunk but poor for viridis. Always check 3+ palettes
- Ignore small-size render: Pretty 160px icon that become blob at 48px is fail
- Forget re-render: Edit function without run build cmd means changes not visible
- Wrong build cmd: Skills use
build-icons.R, agents usebuild-agent-icons.R, teams usebuild-team-icons.R
See Also
- create-glyph — make new glyph from scratch (when enhance not enough)
- audit-icon-pipeline — spot which glyphs need fix across pipeline
- render-icon-pipeline — run full render pipeline after fixes
- ornament-style-mono — visual design rules that apply to glyph composition
- chrysopoeia — value extraction method parallels glyph tune (amplify gold, drop dross)
GitHub репозиторий
Frequently asked questions
What is the enhance-glyph skill?
enhance-glyph is a Claude Skill by pjt222. Skills package instructions and resources that Claude loads on demand, so Claude can perform enhance-glyph-related tasks without extra prompting.
How do I install enhance-glyph?
Use the install commands on this page: add enhance-glyph to Claude Code as a plugin, or clone its repository into your skills directory, then restart Claude so it picks up the skill.
What category does enhance-glyph belong to?
enhance-glyph is in the Meta category, tagged design.
Is enhance-glyph free to use?
Yes. enhance-glyph is listed on AIMCP and free to install. It runs inside Claude, so no separate service account is required to use the skill itself.
Похожие навыки
Этот навык предоставляет проверенную в продакшене настройку для Content Collections — TypeScript-ориентированного инструмента, который преобразует файлы Markdown/MDX в типобезопасные коллекции данных с валидацией Zod. Используйте его при создании блогов, сайтов документации или контентных приложений на Vite + React для обеспечения типобезопасности и автоматической проверки содержимого. Он охватывает всё: от настройки плагина Vite и компиляции MDX до оптимизации развертывания и валидации схем.
Этот навык позволяет разработчикам создавать приложения на платформе прогнозных рынков Polymarket, включая интеграцию с API для торговли и получения рыночных данных. Он также обеспечивает потоковую передачу данных в реальном времени через WebSocket для отслеживания текущих сделок и рыночной активности. Используйте его для реализации торговых стратегий или создания инструментов, обрабатывающих обновления рынка в реальном времени.
Этот навык помогает разработчикам создавать плагины OpenCode, которые подключаются к более чем 25 типам событий, таким как команды, файлы и операции LSP. Он предоставляет структуру плагина, спецификации API событий и шаблоны реализации для модулей на JavaScript/TypeScript. Используйте его, когда вам нужно перехватывать, отслеживать или расширять жизненный цикл ассистента OpenCode AI с помощью пользовательской событийно-ориентированной логики.
SGLang — это высокопроизводительный фреймворк для обслуживания больших языковых моделей (LLM), специализирующийся на быстрой структурированной генерации JSON, regex и рабочих процессов агентов с использованием кэширования префиксов RadixAttention. Он обеспечивает значительно более высокую скорость вывода, особенно для задач с повторяющимися префиксами, что делает его идеальным для сложных структурированных результатов и многократных диалогов. Выбирайте SGLang вместо альтернатив, таких как vLLM, когда вам требуется ограниченное декодирование или вы создаете приложения с интенсивным совместным использованием префиксов.
