CtrlK
BlogDocsLog inGet started
Tessl Logo

vectorization

Guidance for writing and reviewing SIMD / hardware-intrinsics code in dotnet/runtime. USE FOR: vectorizing a scalar algorithm, writing or reviewing code that uses Vector128/Vector256/Vector512, Vector<T>, or the platform intrinsics in System.Runtime.Intrinsics.X86/Arm/Wasm, and validating remainder handling, load/store safety, and hardware-acceleration fallbacks. DO NOT USE FOR: general performance work unrelated to SIMD (use performance-benchmark), or non-vectorized code review (use code-review).

79

Quality

100%

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

100%Weight 40%Scale 1-3

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

The body is a tight, repo-specific enforcement guide: actionable API-level rules, checklists with validation checkpoints, and clearly signaled one-level references. It assumes Claude's competence, adds only dotnet/runtime-specific nuance, and hits the top anchor on every dimension.

DimensionReasoningScore

Conciseness

The body is dense and lean with no padding: every line carries repo-specific nuance (e.g. 'Read IsHardwareAccelerated and Count directly; don't cache them to locals') and it never explains concepts Claude already knows, matching the 'lean and efficient; every token earns its place' anchor rather than the 'could be tightened' (2) level.

3 / 3

Actionability

Concrete, executable guidance throughout — named APIs (Vector128.Create(span), CopyTo, LoadUnsafe(ref T, nuint elementOffset), MemoryMarshal.GetReference, ConditionalSelect, BoundedMemory.Allocate<T>) and exact commands/env vars (DOTNET_EnableAVX2=0, DOTNET_EnableHWIntrinsic=0) — matching 'fully executable... copy-paste ready' rather than the 'pseudocode/missing details' (2) level.

3 / 3

Workflow Clarity

Multi-step processes are clearly sequenced with validation checkpoints: the Testing checklist toggles acceleration via env vars and guards OOB reads with BoundedMemory, and the Review checklist is ordered 'verify in priority order' with an explicit feedback note ('Ask for the missing test rather than just rejecting') — meeting the 'clear sequence with explicit validation steps' anchor rather than 'checkpoints missing/implicit' (2).

3 / 3

Progressive Disclosure

A single well-organized overview file with one-level-deep, clearly signaled references (official .NET SIMD doc, docs/coding-guidelines/vectorization-guidelines.md, .github/copilot-instructions.md, the BoundedMemory source, and PR links); no bundle files exist to require deeper structure, matching the 'clear overview with well-signaled one-level-deep references' anchor rather than 'content that should be separate is inline' (2).

3 / 3

Total

12

/

12

Passed

Description

100%Weight 40%Scale 1-3

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

The description is precise, third-person, and complete: it states what the skill does, gives explicit USE FOR / DO NOT USE FOR triggers with concrete API terms, and disambiguates from sibling skills. It hits the top anchor on every dimension with no fluff or over-claiming.

DimensionReasoningScore

Specificity

Names the domain ('SIMD / hardware-intrinsics code in dotnet/runtime') and lists multiple concrete actions: 'vectorizing a scalar algorithm', 'writing or reviewing code that uses Vector128/Vector256/Vector512, Vector<T>', 'validating remainder handling, load/store safety, and hardware-acceleration fallbacks' — matching the 'lists multiple specific concrete actions' anchor.

3 / 3

Completeness

Explicitly answers both what ('Guidance for writing and reviewing SIMD / hardware-intrinsics code... validating remainder handling, load/store safety') and when via explicit triggers ('USE FOR: ...', 'DO NOT USE FOR: ...'), matching the 'clearly answers both what AND when' anchor rather than the 'when only implied' (2) level.

3 / 3

Trigger Term Quality

Covers the natural terms a dotnet/runtime contributor would say — 'vectorizing a scalar algorithm', 'Vector128/Vector256/Vector512', 'Vector<T>', 'platform intrinsics', 'remainder handling', 'SIMD' — with good breadth; not the 'only some relevant keywords' (2) level and clearly above generic jargon-only (1).

3 / 3

Distinctiveness Conflict Risk

A clear niche (SIMD/vectorization in dotnet/runtime) with an explicit 'DO NOT USE FOR' routing general perf work to performance-benchmark and non-vectorized review to code-review, making wrong-skill triggering unlikely — above the 'could still overlap' (2) level.

3 / 3

Total

12

/

12

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.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 3 suspicious

Warning

Total

15

/

16

Passed

Repository
dotnet/runtime
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.