CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

axiom-swiftui

Use when building, fixing, or improving ANY SwiftUI UI — views, navigation, layout, animations, gestures, debugging, iOS 26 features, iPhone Duo, view-level performance, feature architecture.

Quality

75%

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

Fix and improve this skill with Tessl

tessl review fix ./axiom-cursor/skills/axiom-swiftui/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

63%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-conceived router/index skill with concrete routing signals, explicit conflict-resolution ordering, and a useful anti-rationalization table. Its main weaknesses are structural: the Decision Tree duplicates the Quick Reference table, one section is a contentless stub, and every referenced file is missing from the bundle, so the routing promises cannot be fulfilled as shipped.

Suggestions

Ship the referenced files (e.g., a references/ or skills/ directory containing debugging.md, nav.md, previews.md, etc.) inside the bundle so the router's ~20 'skills/*.md' pointers actually resolve; for cross-suite pointers like 'axiom-macos (skills/swiftui-differences.md)', state where those suites live so navigation still works.

Drop either the Decision Tree or the Quick Reference table — they encode the same symptom-to-file mapping twice; keeping one (the table, with the tree's few unique edges folded in as rows) would cut roughly a quarter of the body.

Expand or remove the 'Cursor Subagent Routing' stub: either list which auditors exist and when delegation is required, or delete the header and fold its one sentence into 'Automated Scanning'.

DimensionReasoningScore

Conciseness

The body avoids concept explanations and assumes Claude's competence, but the Decision Tree digraph duplicates the Quick Reference table's symptom-to-file mapping almost entirely, the 'Cursor Subagent Routing' section is a one-sentence contentless stub, and 'You MUST use this skill for ANY SwiftUI work' repeats the frontmatter description. This is more than 'minor instances' of over-explanation, matching the 'mostly efficient but could be tightened' anchor.

3 / 5

Actionability

Concrete guidance throughout: explicit tool mappings ('xclog launch to axiom_xclog_launch'), named subagents ('swiftui-architecture-auditor'), runnable audit commands ('/axiom-audit swiftui-performance'), and file-plus-section pointers ('skills/debugging.md (Preview Crashes section)'). Not 5 because the routed-to files do not exist in this bundle and the subagent-routing stub is vague about when to delegate.

4 / 5

Workflow Clarity

As a routing skill, the single action (symptom -> reference) is unambiguous via the Quick Reference table, and Conflict Resolution gives an explicit ordered fallback ('Try axiom-swiftui FIRST ... Only use axiom-performance if domain fixes don't help'). Not 5: the 'Cursor Subagent Routing' stub adds confusion, and there is no guidance for symptoms that match no table row or decision-tree edge.

4 / 5

Progressive Disclosure

The router structure is well-signaled with one-level-deep references and section qualifiers, but no bundle files exist in this skill: all ~20 'skills/*.md' references are dangling, and several point outside the bundle entirely ('axiom-macos (skills/swiftui-differences.md)', '/skill axiom-accessibility'). References that cannot be verified to resolve keep this at the 'some structure but could be better organized' level.

3 / 5

Total

14

/

20

Passed

Description

73%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 trigger-optimized description with excellent natural keywords and a clearly scoped SwiftUI niche. Its main gap is that it is all 'when' and no 'what': a reader cannot tell what the skill actually does (routing, reference material, audits) from the description alone.

Suggestions

Add a 'what' clause before the trigger clause, e.g., 'Routes SwiftUI UI work (views, navigation, layout, animations, gestures, debugging, performance) to pattern references, debugging workflows, and automated audits. Use when building, fixing, or improving any SwiftUI UI...'.

Soften 'ANY SwiftUI UI' or scope the performance trigger (e.g., 'view-level performance (see axiom-performance for profiling)') to reduce overlap with the sibling axiom-performance and axiom-uikit suites.

DimensionReasoningScore

Specificity

The verbs are generic ('building, fixing, or improving') but the enumerated domain areas ('views, navigation, layout, animations, gestures, debugging, iOS 26 features, iPhone Duo, view-level performance, feature architecture') are specific and cover the SwiftUI UI space. Matches 'lists several specific actions; minor gaps' — not 5 because the description never states what the skill actually provides, and not 3 because coverage far exceeds '1-2 concrete actions'.

4 / 5

Completeness

The entire description is an explicit 'Use when...' clause, so the 'when' is strong, but the 'what' is never stated — it doesn't say the skill provides patterns, debugging workflows, references, or audits. The topic enumeration conveys scope, lifting it above the 'only when, no what' anchor, but the absent capability statement keeps it below the both-present anchors.

3 / 5

Trigger Term Quality

'SwiftUI', 'navigation', 'layout', 'animations', 'gestures', 'debugging', and 'performance' are exactly the words users say when they need this skill, and 'iOS 26 features', 'iPhone Duo' add distinctive niche triggers. Comprehensive natural-term coverage with no meaningful variation missing.

5 / 5

Distinctiveness Conflict Risk

'ANY SwiftUI UI' carves out a clear niche with distinct triggers, but the aggressive 'ANY' claim plus 'view-level performance' overlaps sibling suites — the body itself concedes a conflict with axiom-performance. 'Mostly distinct; minor overlap risk with closely related skills' is the best fit; not 5 given that acknowledged overlap.

4 / 5

Total

16

/

20

Passed

Validation

100%

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

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

CharlesWiltgen/Axiom

Path

axiom-cursor/skills/axiom-swiftui/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.