CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

axiom-macos

Use when building ANY macOS app — windows, menus, sandboxing, distribution, AppKit bridging or modernization (control events, state restoration, concentric corners), touch input, or macOS-specific SwiftUI patterns.

Quality

73%

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-macos/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-designed router/index skill: a dense symptom-to-reference routing table, explicit cross-suite conflict-resolution rules, and no filler concept explanations. Its main defects are a decision tree that wholly duplicates the routing table, and internal references that point to skills/*.md files absent from the bundle, leaving the router with no resolvable destinations.

Suggestions

Delete the 'Decision Tree' dot graph — it duplicates the Quick Reference table row-for-row at a cost of roughly 20 lines.

Ship the referenced skills/*.md files in the bundle (none are present in references/, scripts/, or assets/) or repoint the routes to files that exist, so the one-level-deep navigation actually resolves.

Give the WWDC/Tech Talk session numbers titles or URLs, and either make the 'Cursor Subagent Routing' section concrete (which subagents, for which table rows) or remove it.

DimensionReasoningScore

Conciseness

The routing table is dense and useful, but the ~20-line dot decision tree duplicates the Quick Reference table almost row-for-row, the bold 'You MUST use this skill for ANY macOS-specific development...' line re-states the frontmatter description, and version/time-sensitive markers ('OS27', '2026-272, 2026-289', 'iOS 27+') plus the two-line, content-free 'Cursor Subagent Routing' section add tokens without adding guidance. Not 4: the duplication is more than a minor instance; not 2: no concept explanations and the table is efficiently packed.

3 / 5

Actionability

Concrete, executable routing guidance: 'Another team's app or app group container access denied without a prompt (OS27)' → skills/sandbox-and-file-access.md, and 'Delegate to the touch-auditor subagent or /axiom-audit touch'. Minor gaps: bare WWDC/Tech Talk session numbers ('2026-272, 2026-289') with no titles or links, and partial doc paths ('/security/app-sandbox'). Not 5: those resource references are not directly usable; not 3: the guidance is specific and immediately actionable as routing, not vague or pseudocode.

4 / 5

Workflow Clarity

The single routing action (task area → reference file) is unambiguous via the symptom table and cross-suite rules ('Use axiom-macos for macOS-only concerns... Use axiom-swiftui for cross-platform SwiftUI'). Not 5: conflict resolution covers only 2 of the ~6 sibling suites, and 'Delegate to the appropriate Cursor subagent when this router calls for a specialized auditor' is stated yet only one table row ever calls for one; not 3: the routing sequence is clear and no destructive/batch operations exist to require validation checkpoints.

4 / 5

Progressive Disclosure

References are well-signaled and exactly one level deep (13 'See skills/*.md' routes in a scannable table), which matches the top anchor's structure — but none of the referenced files (skills/windows.md, skills/menus-and-commands.md, skills/touch.md, etc.) exist in the provided bundle; there is no references/, scripts/, assets/, or skills/ directory, so every navigation target is unresolvable. Not 4/5: the routing structure cannot be verified against actual files and the router points at missing files; not 2: nothing that belongs in separate files is inlined and the signaling is clear.

3 / 5

Total

14

/

20

Passed

Description

70%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 strong, trigger-focused description with explicit and specific 'when' guidance and good natural keyword coverage. Its main weakness is that it never states what the skill does — only when to use it and which topic areas it spans — and a few high-frequency trigger terms for distribution and iOS-compatibility work are absent.

Suggestions

Lead with an explicit capability statement (e.g., 'Provides macOS app development reference and routing for...') before the 'Use when' clause so the 'what' is stated rather than implied by the topic list.

Add the natural trigger terms users say for the distribution and iOS-compatibility areas — 'notarization', 'Developer ID', 'Catalyst', 'Designed for iPad' — to match the coverage of the body's own routing table.

DimensionReasoningScore

Specificity

Names the domain ('building ANY macOS app') and many concrete topic areas ('windows, menus, sandboxing, distribution, AppKit bridging or modernization (control events, state restoration, concentric corners), touch input'), but the only action verb is 'building' — it enumerates topics, not the specific actions of the anchor-4 pattern. Not 2: far more detailed than 'Processes PDF files'; not 4: no list of concrete actions the skill performs.

3 / 5

Completeness

The 'when' is explicit and highly specific ('Use when building ANY macOS app — windows, menus, sandboxing, distribution, ...'), with concrete trigger phrases; the 'what' is conveyed through the enumerated topic areas rather than an explicit capability statement. Not 5: what the skill actually does (reference material, routing, auditing) is never stated as an action; not 3: the 'when' is maximally explicit, not missing or weakly implied.

4 / 5

Trigger Term Quality

Good natural-term coverage — 'macOS app', 'windows', 'menus', 'sandboxing', 'distribution', 'AppKit', 'touch input', 'SwiftUI' are phrases users would actually say. A few natural terms are missing ('notarization', 'Developer ID', 'Catalyst', 'menu bar'), so it falls short of the comprehensive synonym coverage of the anchor-5 example. Not 3: coverage is well beyond 'some relevant keywords'.

4 / 5

Distinctiveness Conflict Risk

'macOS' is a clear niche with distinct triggers, and the qualifier 'macOS-specific SwiftUI patterns' attempts to delimit scope. Minor overlap risk remains with closely related sibling suites (the body itself acknowledges overlap with axiom-swiftui and axiom-security), so it does not reach the minimal-conflict anchor. Not 3: far more distinctive than 'Works with document files'.

4 / 5

Total

15

/

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-macos/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.