skills.rest
JustinChangTW
JustinChangTWCOMMUNITY
Explore all skills by this author →
GitHub

allanyiin-skill-creator-advanced

Creates, audits, and validates AI skill packages with deterministic release gates and eval coverage checks.

Updated Aug 22, 2026
Search Tags:#prompt engineering#skill authoring#benchmark#skill validation#lifecycle management#release gate#eval coverage
One-click install
npx skills add https://github.com/JustinChangTW/ai-skills-core --skill allanyiin-skill-creator-advanced-justinchangtw
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: allanyiin-skill-creator-advanced
Source: https://github.com/JustinChangTW/ai-skills-core/tree/main/skills/01-skill-management/allanyiin-skill-creator-advanced
Command: npx skills add https://github.com/JustinChangTW/ai-skills-core --skill allanyiin-skill-creator-advanced-justinchangtw

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pyyaml, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve? Building and maintaining AI skills often becomes ad-hoc prompt writing with no verification, unclear scope boundaries, or release evidence. This Skill turns skill authoring into a repeatable engineering workflow with mechanical gates, eval coverage audits, and lifecycle governance. ## Core Features & Use Cases - Skill Authoring Workflow: Guides creation, revision, review, split/merge, deprecation, and packaging of skills with structured SKILL.md sections (role, decision boundary, workflow, output contract). - Deterministic Audit Scripts: Runs mechanical checks for structure, semantics, lifecycle state, eval coverage, eval quality, benchmark metadata, wrapper drift, and migration governance. - Release and Stage Gates: Enforces PASS/FAIL/BLOCKED verdicts via release_gate.py and stage_gate.py so incomplete skills cannot be claimed as publish-ready. - Use Case: When a user says "review this skill and make its release gate solid", the Skill audits the folder, adds missing audit scripts, runs the release gate, and reports blocking findings with remediation order. ## Quick Start Ask the AI to review your skill folder and run the release gate to report whether it is publish-ready and what must be fixed first.

Frequently Asked Questions about allanyiin-skill-creator-advanced

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

FAQPage Schema
How do I create a new AI skill with proper structure?▼

Define a SKILL.md with YAML frontmatter containing name and description, then add role, decision boundary, workflow, and output contract sections. The skill's scripts like audit_structure.py mechanically verify the structure and report exact gaps.

How to validate a skill before publishing it?▼

Run release_gate.py against the skill folder to get a PASS, FAIL, or BLOCKED verdict. It chains format, structure, semantics, lifecycle, eval coverage, and benchmark checks, and a FAIL result cannot be overridden by manual checklists.

What eval coverage does a skill need for release?▼

The evals.json file must include cases tagged should-trigger, should-not-trigger, near-miss, overlap-neighbor, happy-path, edge-case, and failure-mode, across zh, en, and mixed languages. The audit_eval_coverage.py script reports any missing tags or languages.

Does this skill require benchmark evidence to publish?▼

Missing benchmark artifacts are a warning by default, not a publish blocker. Benchmarks become mandatory only with --require-benchmark flags or when claiming ROI, live integration quality, or cross-host performance.

When should I not use this skill authoring workflow?▼

Do not use it for one-off prompts, single tool schema documentation, or general writing tasks. It is scoped to creating, revising, reviewing, and gating installable skill packages, not domain tasks like writing articles or building slides.

Related Skills

Explore More Software Engineering Skills→
OlatisunkanmiOlatisunkanmi
8

mine.write-skill

Draft and validate SKILL.md files with guided prompts and repository wiring steps.

Community
Intermediate
zlovexforeverzlovexforever

skill-creator

Generate SKILL.md content and run trigger evals with pass-rate benchmarks.

Community
Advanced
shynlee04shynlee04
51

hivefiver-skill-auditor

Audit local skills for coverage, staleness, and missing triggers in HiveFiver v2.

Community
Intermediate
weside-aiweside-ai

codex-task

Submit tasks directly to Codex (gpt-5-codex) for processing.

Official
Basic