CtrlK
BlogDocsLog inGet started
Tessl Logo

axiom-graphics

Use when working with ANY GPU rendering, Metal, OpenGL migration, shaders, 3D content, RealityKit, AR, USD/USDZ files, or display performance. Covers Metal migration, shader conversion, RealityKit ECS, RealityView, USDKit, ProMotion.

62

Quality

72%

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 ./.claude-plugin/plugins/axiom/skills/axiom-graphics/SKILL.md

The canonical home for this skill is axiom-graphics in CharlesWiltgen/Axiom

SKILL.md
Quality
Evals
Security

Quality

Content

67%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-structured router skill with excellent reference signaling, a genuinely useful anti-rationalization table, and concrete API facts woven into the routing guidance. Its weaknesses are significant redundancy across three parallel routing sections and the fact that every referenced sub-skill file is absent from the bundle being evaluated.

Suggestions

Collapse the triple redundancy: keep the Decision Tree as the single routing map and cut the 'When to Use' bullets or the ~30 Example Invocations down to a handful of representative examples.

Ensure the referenced skills/*.md files actually ship with the skill bundle, or inline the minimal guidance needed for the most common cases so the router is not dead-end navigation.

For the most frequent entry points (e.g. MTKView setup, shader conversion), include one short copy-paste-ready snippet inline so a user can act before opening the referenced file.

DimensionReasoningScore

Conciseness

The same routing map is stated three times: the "When to Use" bullet list, the 21-item Decision Tree, and ~30 "Example Invocations" all encode the identical problem-to-file mapping. Individual lines are terse (no concept over-explanation), but this triple redundancy is more than the "minor instances of over-explanation" of anchor 4 and clearly short of anchor 2's padded prose.

3 / 5

Actionability

Pointers are specific and navigable ("realitykit-ref (Part 10)", "display-performance (Part 12)") and are backed by concrete API facts ("preferredFramesPerSecond = 120", "8.33ms frame budget for 120Hz", "only the C-imported CGToneMapping can" express HAGC). It is below anchor 5 because the file itself contains no copy-paste-ready commands or code, only routing direction.

4 / 5

Workflow Clarity

The Decision Tree sequences routing decisions coherently across 21 cases and the Anti-Rationalization table pre-empts wrong paths. As a read-only router it involves no destructive or batch operations, so the validation cap does not apply; it falls short of anchor 5 because Example Invocations partially duplicates the tree as an unsequenced alternative path with no error-recovery guidance.

4 / 5

Progressive Disclosure

References are one level deep, clearly signaled ("See skills/metal-migration.md"), and each is accompanied by a summary of its contents — structurally at anchor 5. However, none of the referenced skills/*.md files exist in the provided bundle (no references/, scripts/, assets/, or skills/ directories), so the navigation cannot be verified and would break as shipped, dropping it to anchor 4.

4 / 5

Total

15

/

20

Passed

Description

77%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-rich description with an explicit and concrete 'Use when' clause and excellent natural keyword coverage including file extensions. Its main weakness is that the 'what' is expressed as a topic list ("Covers...") rather than concrete actions, which caps specificity and completeness just below top marks.

Suggestions

Replace the "Covers Metal migration, shader conversion..." topic list with action-verb statements, e.g. "Ports OpenGL/DirectX code to Metal, converts GLSL/HLSL shaders to MSL, sets up MTKView render loops, and builds RealityKit 3D/AR scenes."

Soften the over-claim "ANY GPU rendering" to the specific platforms it covers (Apple GPU work) to reduce conflict risk with sibling graphics/game skills.

DimensionReasoningScore

Specificity

"Covers Metal migration, shader conversion, RealityKit ECS, RealityView, USDKit, ProMotion" names concrete topics but uses no action verbs, reading as a topic list rather than a list of things the skill does (e.g. "Ports OpenGL to Metal, converts GLSL shaders"). It is above anchor 2 because migration and shader conversion are semi-concrete, but below anchor 4 because no explicit actions are stated.

3 / 5

Completeness

The explicit "Use when working with..." clause answers 'when' with concrete triggers, and "Covers Metal migration, shader conversion..." answers 'what', but the 'what' is topic-shaped rather than action-shaped. Both halves are present and specific (above anchor 3), yet the what-statement lacks the explicit action phrasing of the anchor-5 example.

4 / 5

Trigger Term Quality

"GPU rendering, Metal, OpenGL migration, shaders, 3D content, RealityKit, AR, USD/USDZ files, display performance... ProMotion" covers the natural terms users would say, including synonyms and file extensions (USD/USDZ). Only trivial omissions (e.g. "visionOS", "120Hz") keep it from being unquestionably comprehensive, matching the anchor-5 example's coverage pattern.

5 / 5

Distinctiveness Conflict Risk

The description carves out a clear Apple GPU/3D niche (Metal, RealityKit, USD/USDZ, ProMotion) with distinct triggers, but the over-claim "ANY GPU rendering" plus the existence of a sibling game-graphics router (acknowledged in the body as axiom-games) creates minor overlap risk. Mostly distinct, not minimal-conflict.

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
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.