CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

commonly-design

Use this skill for ANY visual / UI / design work on Commonly — picking colors, choosing fonts, styling new v2 components, building HTML mocks, designing marketing pages, or auditing whether a change matches the brand. Triggers on tasks involving v2.css tokens, accent / blue / focus ring / SF font, brand mark / logo, design system, design tokens, agent avatars / role tints, pod / chat / inspector / composer styling, marketing landing or login pages, screenshots / demo polish, "make this look better", "design this", "polish UI", "match the brand", "what's the right color/font/spacing/radius". Sources of truth live at `frontend/design-system/` (tokens.css, README.md, brand mark SVG, preview cards) and `frontend/src/v2/v2.css` (production tokens). Read both before recommending or shipping anything.

Quality

93%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

88%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A strong pointer-style skill body: every line carries Commonly-specific rules and exact values, workflows are sequenced with real verification steps (screenshot proof, test re-pinning, browser preview), and referenced files are clearly signaled one level deep. The main improvements are moving dated identity-decision details into a deprecation/history note and trimming the inline system summary that duplicates the README.

Suggestions

Move time-sensitive details like "(Signal, chosen 2026-09-03)" into an 'old patterns' or history section so current guidance stays evergreen, and drop the stale "SF font" trigger from the description to match the body's note that the SF-first stack is gone.

Trim or restructure the "What the system says" summary so it holds only the values needed for quick decisions, leaving full spec content to the README it already points to — this reduces duplication with the referenced source of truth.

DimensionReasoningScore

Conciseness

The body is dense with project-specific facts Claude cannot know (exact hexes, font weights, radii, motion timings) and never explains general concepts, so it mostly earns its tokens. However, time-sensitive details — "(Signal, chosen 2026-09-03)" and the dated identity references — sit outside any 'old patterns'/'deprecated' section, and the "What the system says" section duplicates content the workflow already directs to the README, so it fits 'efficient; minor instances that could be trimmed' rather than the lean anchor at 5.

4 / 5

Actionability

As an instruction-only skill it carries no code, but the guidance is fully concrete: exact values (cobalt `#1d3fd1`, ink `#101828`, radius 4/6, "80ms hover, 120ms card hover, 300ms layout"), exact file paths for every source of truth, an executable rule ("Apply through `var(--v2-*)` — never hardcode hex values"), and a "When in doubt" section prescribing specific behavior for the three common cases (mocks, production v2 code, new component types). This is the copy-paste-equivalent of executable guidance for a design skill.

5 / 5

Workflow Clarity

Multiple clearly sequenced processes with explicit validation checkpoints: the read-first workflow orders files ("Read this FIRST"), shipping requires proof ("the PR carries 1440 and 390 screenshots... load-bearing CSS is re-pinned in `v2-layout-invariants.test.ts`"), and "Preview before merge" plus the drift rule ("If they drift... fix it immediately") provide error-recovery feedback loops. Design work is neither destructive nor batch, so the validation cap does not apply.

5 / 5

Progressive Disclosure

The skill self-describes as a "thin pointer" and delivers: a table maps each referenced file to its purpose, references are one level deep (SKILL.md → repo files), and navigation is easy across nine well-organized sections. The minor gap is the inline "What the system says" summary, which restates spec content that lives in the README/tokens.css — matching 'good structure; most content appropriately placed; minor organization gaps' rather than the fully-split anchor at 5.

4 / 5

Total

18

/

20

Passed

Description

96%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

An exemplary skill description: it states concrete capabilities, gives an explicit and extensive trigger list blending natural user phrasing with codebase-specific terms, and points to the sources of truth. The only weaknesses are a few overly generic trigger phrases ("design this", "polish UI") that could collide with general design skills, and a stale "SF font" trigger the body says has been retired.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — "picking colors, choosing fonts, styling new v2 components, building HTML mocks, designing marketing pages, or auditing whether a change matches the brand" — plus a closing directive ("Read both before recommending or shipping anything"). Coverage of the visual/UI domain is comprehensive with no filler, matching the multiple-specific-actions anchor rather than the minor-gaps anchor at 4.

5 / 5

Completeness

Explicitly answers 'what' ("ANY visual / UI / design work on Commonly" with six enumerated activities) and 'when' ("Triggers on tasks involving..." with an extensive trigger list and quoted natural phrases). Both are explicit and concrete, matching the top anchor; the 4 anchor's 'when could be more explicit' does not apply here.

5 / 5

Trigger Term Quality

Covers natural user phrases ("make this look better", "design this", "polish UI", "match the brand", "what's the right color/font/spacing/radius"), technical synonyms ("accent / blue / focus ring", "brand mark / logo", "design tokens"), and file-level terms ("v2.css tokens", "marketing landing or login pages"). This comprehensively spans both what users say and what the codebase contains, exceeding the good-coverage anchor at 4; only a stale "SF font" term is questionable and it does not reduce natural-term coverage.

5 / 5

Distinctiveness Conflict Risk

Mostly a clear niche — Commonly-specific triggers ("v2.css tokens", "agent avatars / role tints", "pod / chat / inspector / composer styling") are highly distinct — but generic quoted phrases like "design this" and "polish UI" overlap with closely related general design skills (the body itself notes Anthropic's `frontend-design` covers overlapping territory). This is the 'mostly distinct; minor overlap risk with closely related skills' anchor, not the minimal-conflict anchor at 5.

4 / 5

Total

19

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository

Team-Commonly/commonly

Path

fleet-skills/commonly-design/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.