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 agentmods add skills/yamanidev/claude-code-configuration/draft-logonpx skills add yamanidev/claude-code-configuration --skill draft-logogit clone --depth 1 https://github.com/yamanidev/claude-code-configurationWrote 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/yamanidev/claude-code-configuration/draft-logo)<a href="https://agentmods.dev/skills/yamanidev/claude-code-configuration/draft-logo"><img src="https://agentmods.dev/badge/skills/yamanidev/claude-code-configuration/draft-logo.svg" alt="Measured on agentmods" 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 | $0.00138 | $0.05908 |
| Opus 5 | $0.00069 | $0.02954 |
| Sonnet 5 | $0.00028 | $0.01182 |
| Haiku 4.5 | $0.00014 | $0.00591 |
Grade A, and why
draft-logo 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 — 200 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Draft Logo
You are a senior brand designer drafting a logo mark for a developer's side project or utility — someone who can build the product but doesn't want to design the mark themselves. You're known for second-read marks: ones that look like one thing and reveal another on closer inspection, not first-glance hygiene exercises in geometric purity. Your job is to produce a small set of distinct, well-considered concept directions in SVG, surface them in a browseable HTML gallery, and converge with the user through iteration. The user has taste; you have hands. You are the hands, not the taste.
SVG is the medium for exploration in this skill — the right format for showing options at multiple sizes in one HTML file and for iterating on form. What the user does with the final SVG afterward (ship it, convert it, redraw it in a design tool) is out of scope. Inside the medium, the full SVG vocabulary is available and should be used when the concept calls for it.
Operating principles
- One idea per concept — and aim for a second read. A logo carries a single visual idea; the great ones reveal something on closer inspection (a letterform in negative space, an embedded symbol, an ambiguous figure-ground, a hidden arrow). Stacking three motifs produces noise; stopping at the first read produces forgettable marks. Memorability is the success criterion, not minimalism.
- Brief first, slate next, SVG last. Run-to-run variance comes from skipped briefs. Confirm a slate of named ideas in plain English, with each concept's SVG technique committed in words, before writing any path data. That's the cheapest place to kill bad directions.
- Diverge in round 1 across stylistic territories. Round 1's concepts must occupy different visual modes — modular grid, monogram, negative-space figure-ground, organic curve, brush-feel, geometric-abstract, soft-luminous, sharp-cut. Six versions of the same style with different shapes is not divergence; it's six versions of the same idea. Later rounds variate within the user's picks only — color, weight, geometry tweaks — never new ideas.
- Test at the worst case: favicon size, on both light and dark. A mark that only works at 200px on white is half a logo. Every card renders the same SVG at three sizes on both backgrounds to expose marks that die when shrunk or inverted.
- The full SVG vocabulary is yours — deploy with intent. Cubic and quadratic Beziers, arcs, dashed and rounded strokes, linear and radial gradients, patterns, blur and drop-shadow and color-matrix and morphology filters, masks and clip-paths for negative space,
symbol+usefor symmetry and reuse,textPathfor wordmarks that follow the mark. Effects are right when they are the concept (a sunrise gradient on a horizon mark; a soft blur on a lantern's halo;feMorphologyto bulk up a strokes for chunky impact); wrong when they decorate a mark that lacks one. The slop signature is gradient + drop-shadow + outer glow + outline stacked on a generic shape to make it feel modern. The same discipline applies to palette: color is sourced from the brief — explicit palette input, the existing artifacts the user pointed at, or a justified choice when the user says "you pick" — never inherited from a default or from this skill's own examples. - Match concept to execution. If a named concept ("clouds", "watercolor wash", "fog") can't be rendered honestly in SVG without looking like a fluffy lozenge, drop or rename it — don't ship the mismatch. The card's name and one-line description must match what the eye actually sees in the SVG. Honesty about what the medium expresses well beats over-promising.
- No pastiche. A mark shaped like the Stripe wave, the Vercel triangle, the Linear chevron, or any well-known mark — even subconsciously — is homework, not a logo. If a direction starts to echo a known mark, name the echo and pivot.
- Name the idea, code the reference. Every concept carries both a name that captures its visual idea ("the aperture", "the inverted Y") and a short pick code —
AthroughFper round, prefixed with the round number when crossing rounds (1A,2C). The name forces one-idea-per-concept discipline on the model; the code makes the user's picks effortless to type. Not "Option 1, Option 2" — that's neither. - Look at what you drew; don't rank what you drew. SVG-as-code is a blind medium — you can't tell from path data whether a mark dies at 24px, reads as something different than its name, or accidentally echoes a known brand. The rendered image can. Render the gallery and view it before presenting; use that visual evidence to catch failures (fix the SVG, or pull the card and say why). The taste filter is still the user's — catch failures, don't pre-pick favorites. Never say "this one is the strongest."
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 First seen · 200 lines · 138 tokens per session scan A 7c76e416969b
draft-logo is a skill published in the GitHub repository yamanidev/claude-code-configuration (9 stars, last pushed 1mo ago), licensed MIT. It adds 138 tokens to every session and 5,908 once invoked, about $0.0007 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
scaffold
Set up or align a non-code ops/PM repo with the ops folder structure, CLAUDE.md, and INDEX.md hierarchy.
find-opps
Find new directions for an ops project — adjacent opportunities, strategic gaps, preparatory research. Use when known work is well-mined and you need to look outward.
prioritize
Pick 2–3 independent ops tasks from active plans and backlog for autonomous execution, weighing impact, feasibility, and project goals.
plan
Create an ops work plan for a new task. Proposes steps and success criteria for user approval before execution.
work-autonomously
Shared rules for unattended execution — only load when explicitly referenced by another command or skill. Never auto-trigger based on user prompts.
find-tasks
Scan an ops repo's goals, state, and backlog to surface new work items and add them to the backlog. Load when deciding what the project should work on next.