skills.rest
amd
amdOFFICIAL
Explore all skills by this author →
WebsiteGitHub

quark-onnx-eval-runner

Verify Quark ONNX skill routing, planning, artifacts, and recovery checks.

154|27|Updated Jun 19, 2024
Search Tags:#routing#validation#quantization#onnx#skill verification#quark
One-click install
npx skills add https://github.com/amd/Quark --skill quark-onnx-eval-runner
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: quark-onnx-eval-runner
Source: https://github.com/amd/Quark/tree/main/.claude/skills-impl/meta/onnx/quark-onnx-eval-runner
Command: npx skills add https://github.com/amd/Quark --skill quark-onnx-eval-runner

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps maintainers confirm that the Quark ONNX skill family still routes requests correctly, produces valid planning and artifact outputs, and handles common failures in a predictable way before a release or after a skill update.

Core Features & Use Cases

  • Routing Verification: Checks that ONNX prompts are routed to the correct quark-onnx-* skill and never misrouted to torch skills.
  • Planning Validation: Confirms that quantization plans are internally consistent, match the requested preset, and respect deployment-target constraints.
  • Artifact and Recovery Checks: Validates manifests and generated scripts, and tests that error cases are diagnosed with practical fixes or proper handoffs.
  • Use Case: A maintainer runs this Skill after changing ONNX workflow wording or presets to catch broken routing, invalid manifests, or unsupported preset combinations before shipping.

Quick Start

Ask the skill to manually verify the Quark ONNX skill family across routing, planning, artifact, and recovery cases and record the results in a validation report.

Frequently Asked Questions about quark-onnx-eval-runner

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I validate ONNX quantization plans before a release?▼

To validate ONNX quantization plans before a release, run schema-aware checks that confirm internal consistency, match requested presets, and respect deployment-target constraints. This generates a validation report confirming your plans are correct and ready for deployment.

How do I verify ONNX skill routing and prevent misrouting to torch workflows?▼

Verifying ONNX skill routing requires checking that prompts route to the correct quark-onnx-* skill and never misroute to torch skills. This cross-backend verification enforces correct request routing before shipping workflow updates.

Does manual evidence collection work for testing ONNX error recovery scenarios?▼

Manual evidence collection works for testing ONNX error recovery by validating that error cases are diagnosed with practical fixes or proper handoffs. This approach confirms your skill handles common failures predictably after an update.

What is the best way to check ONNX manifests and generated scripts for deployment?▼

The best way to check ONNX manifests and generated scripts is through artifact validation that tests manifests and scripts against schema-aware rules. This confirms correct diagnosis of ONNX error scenarios and validates artifacts for deployment targets.

Related Skills

Explore More Software Engineering Skills→
amdamd
154

quark-onnx-doc-drift-check

Compare Quark ONNX skill guidance against upstream docs and source.

Official
Advanced
eins78eins78
2

plot-release

Generate RC checklists and release notes for Plot and git-based releases.

Community
Advanced
amdamd
154

quark-onnx-router

Routes Quark ONNX user requests to the smallest correct downstream workflow.

Official
Advanced
amdamd
154

quark-onnx-ptq

Execute end-to-end post-training quantization workflows for ONNX models with AMD Quark.

Official
Advanced