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 sumitttt4/Glyph-Skill --skill glyphgit clone --depth 1 https://github.com/sumitttt4/Glyph-SkillWrote 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/sumitttt4/glyph-skill/glyph)<a href="https://agentmods.dev/skills/sumitttt4/glyph-skill/glyph"><img src="https://agentmods.dev/badge/skills/sumitttt4/glyph-skill/glyph/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/sumitttt4/glyph-skill/glyph"><img src="https://agentmods.dev/badge/skills/sumitttt4/glyph-skill/glyph.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 | $0.00037 | $0.01457 |
| Opus 5.5 | $0.00015 | $0.00583 |
| Sonnet 5.5 | $0.00007 | $0.00291 |
| Haiku 4.5 | $0.00004 | $0.00146 |
Grade A, and why
glyph 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 yesterday.
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 — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Glyph — frontend design judgment
One skill, one entry point. Help the assistant decide, compose, build, interact, and verify. Do not reproduce the Glyph website as a template.
Instruction roles
- Requirements protect functionality, truthful claims and accessibility. State conflicts and limitations.
- Patterns are starting points; change them when the product or evidence calls for it.
- Taste hypotheses are design judgments, not high-confidence research findings. Explain and inspect them.
- The user's goals and existing brand take precedence over aesthetic defaults. Never claim this skill overrides system instructions. Do not brand an unnamed product “Glyph”; use a descriptive working title and label it provisional.
Workflow
- Understand. Inspect existing UI and code before edits. Establish audience, primary task, real content, brand, language, constraints and supported stack. Ask only for facts that materially block progress. Preserve unrelated work.
- Choose. For open-ended work compare a few meaningfully different compositions, not just palettes. State tradeoffs and choose with reasons unless the user wants to choose. Preserve established direction unless redesign is requested. Read direction and assets. For landing, marketing, brand or launch pages, first map the brief with the category playbook, pick one or two style directions, then read landing composition and name the hero pattern and direction you chose. Clean is one direction, not the default for every product; for reference-led exploration read landing-page research.
- Compose. Sketch or code the important screen with actual content. Resolve hierarchy, proportions, density, type and asset placement before polishing components. Figma is optional. Check the primary task at small and large sizes.
- Express. Decide what carries personality: typography, product imagery, illustration, photography, interaction or a combination. Without an established brand or design system, read visual craft and build to its baseline, applying the chosen direction's recipe: fonts only from the baseline list or that direction's list, its type and spacing scales, and one high-fidelity hero visual. Never improvise fonts or hand-drawn object illustrations. Verify asset provenance and rights. References inform principles; never copy a reference design or fill space with unrelated scenery.
- Build. Read implementation and relevant topic references below. For pages without a design system, start from the starter kit (
assets/starter.css) and its direction presets rather than inventing tokens. Use the actual project stack. Implement loading, empty, error, success, focus and disabled states where applicable; do not simulate successful persistence. - Interact. Read motion for interactions that benefit from movement. Decide whether to animate at all. Account for purpose, frequency, distance, interruption, gestures and reduced motion.
- Verify and revise. Read verification. Inspect real renders and exercise interactions. For visual work, run the craft checklist in visual craft at desktop and mobile widths. Compare with the chosen direction and revise. Report checks performed, failures, limitations and remaining risks. Never equate a successful build with a successful interface.
What ships with it
23 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.
- assets/starter.css 17 KB
- LICENSE 1.0 KB
- MIGRATION.md 3.8 KB
- references/accessibility-rules.md 2.5 KB
- references/anti-slop-rules.md 2.4 KB
- references/background-rules.md 1.7 KB
- references/category-playbook.md 10 KB
- references/component-rules.md 4.8 KB
- references/core-color-rules.md 2.4 KB
- references/direction-composition-assets.md 2.5 KB
- references/framework-rules.md 1.8 KB
- references/implementation.md 2.4 KB
- references/landing-composition.md 4.4 KB
- references/landing-page-research.md 4.1 KB
- references/layout-rules.md 2.3 KB
- references/motion-rules.md 3.3 KB
- references/responsive-rules.md 2.3 KB
- references/starter-kit.md 5.3 KB
- references/style-directions.md 12 KB
- references/typography-rules.md 2.4 KB
- references/verification.md 2.8 KB
- references/visual-craft.md 11 KB
- THIRD_PARTY_NOTICES.md 2.9 KB
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.
- yesterday Changed · +5 lines 49a292929d88
- 11d ago First seen · 61 lines · 37 tokens per session scan A 986e27088003
glyph is a skill published in the GitHub repository sumitttt4/Glyph-Skill (14 stars, last pushed 2d ago), licensed MIT. It adds 37 tokens to every session and 1,457 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-10-26.
glyph compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| glyphsumitttt4/Glyph-Skill | Skill | 14 | 2d ago | MIT | A | 37 | Claude Code, Codex |
| anti-slop-designsickn33/agentic-awesome-skills | Skill | 49k | yesterday | MIT | A | 34 | Claude Code |
| chakra-ui-builderchakra-ui/chakra-ui | Skill | 41k | yesterday | MIT | A | 214 | Claude Code, Codex |
| visual-ralphYeachan-Heo/oh-my-codex | Skill | 34k | 3d ago | MIT | A | 52 | Codex |
| open-design-setupcompozy/compozy | Skill | 2.8k | yesterday | MIT | A | 67 | Claude Code, Codex |
| frontend-visual-qadaymade/claude-code-skills | Skill | 1.5k | yesterday | MIT | A | 125 | Claude Code, Codex |
| prototype-webnexu-io/html-anything | Skill | 9.4k | 24d ago | Apache-2.0 | A | 24 | Claude Code, Codex |