multi-ai-verification

Multi-layer quality assurance with 5-layer verification pyramid (Rules → Functional → Visual → Integration → Quality Scoring). Independent verification with LLM-as-judge and Agent-as-a-Judge patterns. Score 0-100 with ≥90 threshold. Use when verifying code quality, security scanning, preventing test gaming, comprehensive QA, or ensuring production readiness through multi-layer validation.

Quality

73%

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/multi-ai-verification/SKILL.md

Quality

Content

62%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 body delivers a well-sequenced, highly actionable 5-layer workflow with explicit validation gates and feedback loops. Its main weaknesses are verbosity (padding, illustrative report templates, restated content) and a monolithic inline structure that should split reference material into separate files.

Suggestions

Replace the full illustrative Layer 1-5 example reports (with fake checkmarks/numbers) with brief one-line report skeletons; Claude can populate the details without ~150 lines of sample output.

Remove explanations of tools Claude already knows (the 'Tools Available' list and the duplicate Appendix C catalog) and state only the verification-specific expectations.

Move the full operational scoring rubrics (Appendix B) and tool reference (Appendix C) into separate reference files and link to them from the body, reducing the monolithic inline bulk.

Dimension	Reasoning	Score
Conciseness	The ~1000-line body is noticeably verbose: full illustrative Layer 1-5 report templates with fake checkmarks and numbers, explanations of tools Claude already knows ('Linters (ESLint, Pylint)', 'Type checkers (TypeScript, mypy)'), the pyramid restated four times (ASCII art, table, gates, quick reference), and a duplicate tool catalog in Appendix C.	2 / 5
Actionability	Provides concrete executable commands throughout ('npx eslint src/*/.{ts,tsx,js,jsx}', 'npx tsc --noEmit', 'npx semgrep --config=auto src/', 'bandit -r src/', 'npx playwright test --screenshot=on') and real task() prompts, with only minor illustrative-only report blocks keeping it below 5.	4 / 5
Workflow Clarity	The 5-layer workflow is clearly sequenced with each operation having a numbered Process, an explicit Validation checklist, and a pass/fail Gate, plus a feedback loop ('Failed Gate → Gap Analysis → Apply Fixes → Re-Verify → Repeat Until Pass') for error recovery.	5 / 5
Progressive Disclosure	No bundle files exist and the entire ~1000-line document is inline; it has genuine section/appendix structure, but content that belongs in separate files (Appendix B operational rubrics, Appendix C tool reference, full example reports) is inlined with no external references signaled.	3 / 5
	Total	14 / 20 Passed

Description

83%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 and complete: it clearly states what the skill does and when to use it with concrete trigger phrases and a distinctive verification framing. Its only weakness is that the listed actions are somewhat high-level and overlap with adjacent review/security skills.

Dimension	Reasoning	Score
Specificity	Lists several specific actions ('verifying code quality, security scanning, preventing test gaming, comprehensive QA, production readiness') and concrete patterns (LLM-as-judge, Agent-as-a-Judge, 0-100/≥90 scoring), but the actions are high-level categories rather than granular operations, so it sits below the comprehensive 5 anchor.	4 / 5
Completeness	Explicitly answers both 'what' (5-layer verification pyramid, independent LLM-as-judge/Agent-as-a-Judge, 0-100 with ≥90 threshold) and 'when' ('Use when verifying code quality, security scanning...') with concrete trigger phrases.	5 / 5
Trigger Term Quality	Includes natural phrases users would say ('code quality', 'security scanning', 'comprehensive QA', 'production readiness') with good coverage, but lacks synonyms/variations that would push it to a 5.	4 / 5
Distinctiveness Conflict Risk	The 5-layer pyramid + LLM-as-judge framing is a clear niche, but trigger terms like 'security scanning' and 'code quality' risk overlap with dedicated security-review and code-review skills, keeping it just below the minimal-conflict 5 anchor.	4 / 5
	Total	17 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (1010 lines); consider splitting into references/ and linking	Warning
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning

	Total	14 / 16 Passed

Repository: fernandezbaptiste/Skrillz
Path: skills/multi-ai-verification/SKILL.md
Commit: 93ed392

Reviewed: 2 days 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.