hqq-quantization

Half-Quadratic Quantization for LLMs without calibration data. Use when quantizing models to 4/3/2-bit precision without needing calibration datasets, for fast quantization workflows, or when deploying with vLLM or HuggingFace Transformers.

Quality

92%

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

Quality

Content

85%

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

A highly actionable, well-structured skill body with executable code, sequenced workflows, validation steps, and clean progressive disclosure into real reference files. Its only weakness is length and some repeated config examples that could be consolidated.

Suggestions

Consolidate the repeated BaseQuantizeConfig/HqqConfig blocks — the basic config is shown nearly verbatim in 'Basic quantization', 'Quantization configuration', 'Quantize full model', and Workflow 1; show one canonical example and reference it.

Trim the 'Key advantages' bullet list, which restates the opening sentence and the 'When to use HQQ' list, to reduce token overhead.

Consider moving the full backend list and selection table into references/advanced-usage.md, keeping only the most common backends inline.

Dimension	Reasoning	Score
Conciseness	Assumes Claude's competence and never explains basic concepts, but the ~440-line body repeats BaseQuantizeConfig/HqqConfig usage multiple times and the 'Key advantages' bullets restate the opening and 'When to use' list, leaving room to tighten.	2 / 3
Actionability	Provides fully executable, copy-paste-ready code throughout — pip install variants, BaseQuantizeConfig/HqqConfig, vLLM LLM setup, and PEFT LoraConfig — with specific parameter values.	3 / 3
Workflow Clarity	Workflow sections are explicitly numbered (1. Configure, 2. Load and quantize, 3. Verify quality, 4. Save) and include a generation-based validation checkpoint plus error→fix patterns in Common issues.	3 / 3
Progressive Disclosure	SKILL.md is an overview with clearly signaled one-level-deep markdown links to references/advanced-usage.md and references/troubleshooting.md; both files exist and contain no further nested references.	3 / 3
	Total	11 / 12 Passed

Description

100%

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 concise, well-targeted description that states concrete capabilities and provides explicit use-when triggers with natural vocabulary and a clear, conflict-resistant niche.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'Half-Quadratic Quantization for LLMs without calibration data', 'quantizing models to 4/3/2-bit precision', 'deploying with vLLM or HuggingFace Transformers' — in third person voice.	3 / 3
Completeness	Explicitly answers both what (calibration-free HQQ for LLMs) and when via a clear 'Use when ...' trigger clause covering precision levels and deployment targets.	3 / 3
Trigger Term Quality	Covers natural terms users would say: 'quantizing models', '4/3/2-bit precision', 'calibration datasets', 'fast quantization', 'vLLM', 'HuggingFace Transformers'.	3 / 3
Distinctiveness Conflict Risk	Occupies a clear niche (calibration-free quantization) with specific framework triggers, distinguishing it from GPTQ/AWQ/bitsandbytes alternatives and making misfiring unlikely.	3 / 3
	Total	12 / 12 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: synthetic-sciences/openscience
Path: backend/cli/skills/ml-training/hqq/SKILL.md
Commit: 52845c3

Reviewed: about 13 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.