add-uint-support

Add unsigned integer (uint) type support to PyTorch operators by updating AT_DISPATCH macros. Use when adding support for uint16, uint32, uint64 types to operators, kernels, or when user mentions enabling unsigned types, barebones unsigned types, or uint support.

Quality

78%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills_all/add-uint-support/SKILL.md

Quality

Content

75%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.

The content is highly actionable and well-organized, with concrete before/after code, a decision tree, and a validation checklist. Its main weakness is redundancy — the Method 1/Method 2 examples and the step sequence are repeated across several sections, which inflates length without adding guidance — and a lack of an explicit error-recovery feedback loop.

Suggestions

Collapse the duplicate Method 1/Method 2 before/after examples: show them once in Step 4 and have Quick reference, Common patterns, and Multiple dispatch sites refer back or show only the minimal delta.

Merge the trailing Workflow section into the Instructions steps (or replace one with the other) so the sequenced procedure and its validation checklist live in a single place.

Add an explicit validation feedback loop (e.g., build/compile after edits; if dispatch errors appear, re-check that every dispatch site got AT_EXPAND(AT_BAREBONES_UNSIGNED_TYPES) or AT_INTEGRAL_TYPES_V2 and retry) rather than only a static checklist and deferring all testing to the user.

Dimension	Reasoning	Score
Conciseness	The body assumes Claude's competence (no basic-concept padding) and is mostly concrete, but the same Method 1/Method 2 before/after patterns recur across Quick reference, Step 4, Common patterns, and Multiple dispatch sites, and the trailing Workflow section restates the Instructions steps — clear tightening opportunities.	3 / 5
Actionability	Guidance is fully executable: concrete before/after C++ snippets with exact macro names (AT_BAREBONES_UNSIGNED_TYPES, AT_INTEGRAL_TYPES_V2), a decision tree, and worked multi-site examples give copy-paste-ready direction for the common cases.	5 / 5
Workflow Clarity	Steps 1-7 are clearly sequenced with a validation checklist in Step 7 and a decision tree, but there is no explicit validate-then-fix-then-retry feedback loop, and the separate Workflow section omits the validation step present in Instructions — a minor checkpoint gap.	4 / 5
Progressive Disclosure	The single-file skill is well structured with clear ##/### headings (Quick reference, Instructions, Common patterns, Edge cases, Decision tree) and no nested references; at ~320 lines some worked-example sections could be split out, but organization and navigation are good.	4 / 5
	Total	16 / 20 Passed

Description

82%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.

The description is strong: it pairs a concrete statement of what the skill does with an explicit, trigger-rich "Use when" clause and a clearly scoped niche. Its only weakness is moderate action specificity, since the core operation is essentially a single macro-edit pattern rather than a list of distinct capabilities.

Dimension	Reasoning	Score
Specificity	The description names the domain (PyTorch operators) and one concrete action ("updating AT_DISPATCH macros") to add uint type support, but does not enumerate multiple distinct actions, matching the anchor for 1-2 concrete actions without comprehensive coverage.	3 / 5
Completeness	It explicitly answers both what ("Add unsigned integer (uint) type support to PyTorch operators by updating AT_DISPATCH macros") and when ("Use when adding support for uint16, uint32, uint64 types ... or when user mentions ... uint support") with concrete trigger phrases.	5 / 5
Trigger Term Quality	It covers natural terms a user would say — "uint support", "unsigned types", "barebones unsigned types", "uint16, uint32, uint64", "operators, kernels" — giving good keyword coverage with synonyms, though a couple of natural phrasings are absent.	4 / 5
Distinctiveness Conflict Risk	The niche is sharply defined (unsigned integer types via AT_DISPATCH macros for PyTorch operators) with distinctive triggers that are unlikely to fire for unrelated skills, minimizing conflict risk.	5 / 5
	Total	17 / 20 Passed

Validation

100%

Warnings & errors only

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: Microck/ordinary-claude-skills
Path: skills_all/add-uint-support/SKILL.md
Commit: 1056d29

Reviewed: ~9 hours ago

Table of Contents

Discovery Implementation Validation

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.