add-uint-support
Add Unsigned Integer Support to PyTorch Operators
PyTorch operators may reject uint16, uint32, and uint64 tensors when dispatch macros omit unsigned types. This skill guides precise AT_DISPATCH_V2 updates for complete unsigned type coverage.
Install with my Agent
Copy this request to your Agent. It includes the canonical Skill page and manifest.
Review the Skillstore skill "add-uint-support" from https://skillstore.io/skills/pytorch-add-uint-support.md and its manifest at https://skillstore.io/api/skills/pytorch-add-uint-support/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.Your Agent should still show its plan and request any confirmation required by the security policy.
Agent-readable resources
Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.
Test it
Using "add-uint-support". Add unsigned support to an operator that dispatches all standard types.
Expected outcome:
The dispatch gains the barebones unsigned type group while preserving existing floating and signed integer coverage.
Using "add-uint-support". Update an operator that separately dispatches integral and floating types.
Expected outcome:
The signed integral group is replaced with its uint-capable superset. The floating group remains unchanged.
Using "add-uint-support". Review a file that may already support unsigned tensors.
Expected outcome:
The review finds existing uint coverage and recommends no dispatch change, avoiding duplicate type entries.
Security Audit
SafeAll 38 external-command findings are false positives caused by Markdown backticks and fenced C++ examples. The two reconnaissance findings are ordinary C++ function declarations, and no malicious intent or prompt injection was found.
Risk Factors
⚙️ External commands (38)
Share & cite this report
Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.
Copy report link
https://skillstore.io/skills/pytorch-add-uint-support/audits/8?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown badge
[](https://skillstore.io/skills/pytorch-add-uint-support?utm_source=security_passport_badge)HTML badge
<a href="https://skillstore.io/skills/pytorch-add-uint-support?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/pytorch-add-uint-support/security.svg" alt="Skillstore security assessment" loading="lazy"></a>Embed card
<iframe src="https://skillstore.io/embed/skills/pytorch-add-uint-support.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>Academic citations (APA · BibTeX · CFF)
APA citation
pytorch. (2026). add-uint-support security audit report (audit version 8) [Author version unspecified]. Skillstore. https://skillstore.io/skills/pytorch-add-uint-support/audits/8BibTeX citation
@techreport{pytorch-pytorch-add-uint-support-2026,
author = {pytorch},
title = {add-uint-support security audit report (audit version 8)},
institution = {Skillstore},
year = {2026},
number = {8},
url = {https://skillstore.io/skills/pytorch-add-uint-support/audits/8},
note = {Author version unspecified}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "add-uint-support security audit report (audit version 8)"
version: "unspecified"
type: report
authors:
- name: "pytorch"
date-released: "2026-07-23"
url: "https://skillstore.io/skills/pytorch-add-uint-support/audits/8"
identifiers:
- type: other
value: "skillstore:pytorch-add-uint-support:audit:8"
description: "Skillstore immutable audit report identifier"
Skillstore Score
Why this score Evidence Confidence: HighWhat You Can Build
Extend an operator
Add uint16, uint32, and uint64 dispatch coverage to an existing PyTorch operator.
Align kernel dispatch sites
Apply the same unsigned type coverage across CPU, CUDA, and helper kernels.
Review uint changes
Check whether a proposed dispatch update uses the correct type group without duplicate coverage.
Try These Prompts
Add uint16, uint32, and uint64 support to this PyTorch operator. Explain each dispatch change.
Find AT_INTEGRAL_TYPES dispatch entries in this file and replace them with the appropriate uint-capable group.
Inspect all CPU and CUDA dispatch sites in this operator. Add consistent unsigned coverage and identify any sites left unchanged.
Analyze this operator's dispatch groups, existing uint coverage, and semantic constraints. Propose minimal updates and a focused test plan.
Best Practices
- Inspect every dispatch site before editing the operator.
- Prefer AT_INTEGRAL_TYPES_V2 when the signed integral group is already present.
- Compile and test uint16, uint32, and uint64 behavior after each change.
Avoid
- Do not add unsigned types to operators that only support floating values.
- Do not combine duplicate unsigned groups in one dispatch list.
- Do not assume successful compilation proves correct unsigned arithmetic.