nwave-ai/nw-quality-validation New SkillScore Maintained Type-specific validation checklists, six quality characteristics, and quality gate thresholds for documentation assessment
Documentation by nwave-ai
Use this skill Links this skill to your agents and keeps it current with the author’s updates.
On-demand On-demand: invoked explicitly — never occupies your agent's context until called.
◆ Doc Ops pack
1 · One-time — install the CLI (
docs ↗ )
$ pip install decimalaiCopy
2 · Get the skill — writes .claude/skills/nwave-ai-nw-quality-validation/SKILL.md
$ decimalai skills pull nwave-ai-nw-quality-validation --out .claude/skills/Copy
Copy link Copy README badge Copy SKILL.md No CLI? Get the same file — curl · raw file $ curl -fsSL --create-dirs -o .claude/skills/nwave-ai-nw-quality-validation/SKILL.md https://app.decimal.ai/s/nwave-ai-nw-quality-validation/SKILL.mdCopy
how this skill behaves on real traffic · last 30 days
No production data for this skill yet — usage numbers appear here once agents start running it.
Anonymized aggregate across every org using this skill — proof from real production use, not a demo. Sign in and install to see your own agents here →
· Benchmark: with vs. without sign in to run this on your agents 22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 2h ago
— -14% tokens 0% turns 95% of whole cases pass with skill (21/22)
Showing 5 of 22 cases See all cases & grading →
SkillSafety ✓ 3 of 3 passed ✓ Code scan 15 of 15 checks clean
0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026
live_secret remote_code_execution reverse_shell data_exfiltration ssrf_metadata agent_snooping anti_refusal destructive_commands obfuscated_unicode unicode_homoglyph opaque_blob suspicious_url tool_overreach trigger_abuse prompt_injection_phrasing
✓ Intent Safe
LLM judge · does the skill do what it claims
✓ Content Clean
LLM judge · policy & harmful-content review
Source Imported License MITImported 8/3/2026Version v1
Contents Benchmark Live Traces Versions (1) Ratings (0)
Quality Validation Six Quality Characteristics Accuracy : Factually correct, technically sound, current | Validation: expert review, automated testingCompleteness : All necessary info present for doc type | Validation: checklist, gap analysisClarity : Easy to understand, logical flow, appropriate level | Validation: Flesch 70-80Consistency : Uniform terminology, formatting, structure | Validation: style guide complianceCorrectness : Proper grammar, spelling, punctuation | Validation: automated check, zero errorsUsability : User achieves goal efficiently | Validation: task success, DIVIO type purpose servedQuality Gate Thresholds | Metric | Threshold | |--------|-----------| | Readability (Flesch) | 70-80 | | Spelling errors | 0 | | Broken links | 0 | | Style compliance | 95%+ | | Type purity | 80%+ single type |
Type-Specific Validation Checklists Tutorial Checklist ] Completable without external references | ] Steps numbered and sequential ] Each step has verifiable outcome | ] No assumed prior knowledge | ] Builds confidence How-to Checklist ] Clear goal stated upfront | ] Assumes fundamentals known ] Single task focus | ] Ends with completion indicator | ] No basics teaching Reference Checklist ] All parameters documented | ] Return values specified ] Error conditions listed | ] Examples per entry | ] No narrative Explanation Checklist ] Addresses "why" not "what" | ] Provides context and reasoning ] Discusses alternatives | ] No task steps | ] Builds conceptual model Verdict Criteria approved : Passes all type-specific validation, no collapse violations, meets quality gatesneeds-revision : Minor issues fixable in place (clarity, missing examples, small gaps)restructure-required : Collapse detected requiring split, or fundamental type mismatchWant to see if nwave-ai/nw-quality-validation would activate on your phrasing? Try it before you install — no LLM call, no signup. Try this skill →