CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

impeccable

Use when the user wants to design, redesign, shape, critique, audit, polish, clarify, distill, harden, optimize, adapt, animate, colorize, extract, or otherwise improve a frontend interface. Covers websites, landing pages, dashboards, product UI, app shells, components, forms, settings, onboarding, and empty states. Handles UX review, visual hierarchy, information architecture, cognitive load, accessibility, performance, responsive behavior, theming, anti-patterns, typography, fonts, spacing, layout, alignment, color, motion, micro-interactions, UX copy, error states, edge cases, i18n, and reusable design systems or tokens. Also use for bland designs that need to become bolder or more delightful, loud designs that should become quieter, live browser iteration on UI elements, or ambitious visual effects that should feel technically extraordinary. Not for backend-only or non-UI tasks.

Quality

88%

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

The canonical home for this skill is impeccable in pbakaus/impeccable

SKILL.md
Quality
Evals
Security

Quality

Content

77%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 well-orchestrated router body: concrete commands, explicit bounded verification, clear routing and recovery paths, and disciplined token economy. Its major defect is that the entire disclosed detail layer — all referenced reference/*.md files carrying the actual design guidance — is missing from the bundle, breaking navigation and leaving the core instruction content unverifiable.

Suggestions

Ship the reference/ bundle: all ~26 linked paths (reference/new-work.md, reference/craft-floor.md, reference/routing.md, and every per-command reference in the Commands table) are absent from the provided bundle, so every navigation link in the body dead-ends.

State the 'craft is a deprecated alias for ordinary new-work' note once — either the Commands-table row or the Routing bullet — instead of duplicating it in both places.

Split the overloaded Setup step 1 into the core instruction (run `scripts/impeccable context` once per session) plus a short fallback note covering the Windows `.cmd` launcher and the no-base-dir path, so the primary command is scannable.

DimensionReasoningScore

Conciseness

The body is lean and assumes competence — no space is spent explaining design concepts Claude already knows — but there is trimmable redundancy: "craft is a deprecated alias for ordinary new-work" is stated both in the Commands table row and again in Routing ("`craft` is a deprecated alias for ordinary new-work and adds nothing"), and Setup step 1 buries the core command in a long chain of qualifications. This is the level-4 anchor ('minor instances ... that could be trimmed'), not level 5, because of that duplication and the overloaded setup sentence.

4 / 5

Actionability

Concrete, executable commands are present throughout — `scripts/impeccable context`, `impeccable pin <pin|unpin> <command>`, `$impeccable hooks <on|off|status|...>`, `$impeccable doctor`, plus a 23-row Commands table and explicit routing rules and a launcher-failure fallback. It falls short of level 5 because the substantive design guidance for the common cases is entirely delegated to reference files ("load its reference ... and follow it") that are not present in the provided bundle, so the executable chain dead-ends in practice.

4 / 5

Workflow Clarity

Setup is a clear numbered sequence (1–3) with an explicit recovery path for launcher failure ("send a separate message before the next tool call ... read existing PRODUCT.md and DESIGN.md"), and verification is an explicit bounded feedback loop ("Build fully, inspect once with a batched round ... fix everything it shows in one batch, confirm with at most one more round, and stop polishing"). Edge-case sequencing is also handled (resume without rerunning `impeccable context` after init; `CONTEXT_STALE` handling). This matches the level-5 anchor: clear sequence, explicit validation steps, and feedback loops for error recovery.

5 / 5

Progressive Disclosure

The body's structure is textbook one-level-deep disclosure — a Commands table mapping every command to a named reference, plus clearly signaled pointers like "read [routing.md](reference/routing.md) and present its context-aware menu" — but scoring against the actual bundle, none of the ~26 referenced `reference/*.md` files exist in the provided bundle (only `scripts/` ships). Every navigation link dead-ends, so the otherwise well-signaled structure is not usable as-is; this sits below level 4's 'references mostly clear' and at level 3's 'some structure but could be better organized'.

3 / 5

Total

16

/

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.

A dense, trigger-rich description that explicitly answers both what and when with concrete verbs, surfaces, and quality dimensions. The only weakness is minor trigger overlap with adjacent skills (dashboards/dataviz, optimize/extract) despite the explicit not-for clause.

DimensionReasoningScore

Specificity

Lists many concrete actions ("design, redesign, shape, critique, audit, polish, clarify, distill, harden, optimize, adapt, animate, colorize, extract") tied to concrete surface types ("websites, landing pages, dashboards, product UI, app shells, components, forms, settings, onboarding, and empty states") plus concrete aspect areas ("typography, fonts, spacing, layout, alignment, color, motion, micro-interactions, UX copy"). This matches the level-5 anchor of comprehensive, specific concrete actions; nothing sits at the level-4 'minor gaps' rung because coverage spans actions, artifacts, and quality dimensions.

5 / 5

Completeness

The 'when' is explicit and trigger-anchored ("Use when the user wants to design, redesign ... or otherwise improve a frontend interface") and the 'what' is concrete ("Handles UX review, visual hierarchy, information architecture, cognitive load, accessibility, performance, responsive behavior, theming"). Both questions are answered explicitly with concrete trigger phrases, matching the level-5 anchor; the level-4 anchor's 'when could be more explicit' caveat does not apply.

5 / 5

Trigger Term Quality

Natural user phrasings are comprehensively covered, including synonyms and scenario language users actually say: "critique", "audit", "polish", "bland designs that need to become bolder", "loud designs that should become quieter", "live browser iteration on UI elements". Even at the level-4 anchor I cannot name a common natural term that is missing; frontend work has no meaningful file-extension trigger to add.

5 / 5

Distinctiveness Conflict Risk

The frontend-UI niche is clear and the explicit boundary "Not for backend-only or non-UI tasks" narrows conflicts, but several scoped terms carry minor overlap with closely related skills: "dashboards" overlaps data-visualization skills, "optimize" and "extract" overlap general performance-tuning and data-extraction skills. This lands on the level-4 anchor ('minor overlap risk with closely related skills') rather than level 5's 'minimal conflict risk'.

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

relative_links

Relative link issues: 34 missing

Warning

Total

15

/

16

Passed

Repository

Raudbjorn/IRONMAW

Path

.agents/skills/impeccable/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.