managing-skills

스타12

포크1

업데이트2026년 7월 22일 06:38

Audits and maintains a local skill collection, including discovery, Codex compatibility, trigger scope, installation paths, bundled resources, and validation. Use for "스킬 분석", "스킬 현황", "스킬 점검", "스킬 개선", or skill ecosystem maintenance. Use the built-in skill creator for new skills and substantial rewrites.

설치

Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.

출처

jiunbae

jiunbae/agent-skills

GitHub 저장소 열기 Creator 저장소 보기

다운로드

Skill Manager

Maintain a skill repository without loading or installing every skill by default.

Workflow

Identify the source-of-truth repository and synchronize it only when the working tree is clean or the update can be proven safe.
Inventory SKILL.md files, installed links, and the skills exposed in the current session. Treat those as separate sets.
Run the current Codex validator for every skill directory.
Review descriptions for precise positive triggers and important non-trigger boundaries. Broad descriptions cause unwanted implicit invocation.
Check folder/name consistency, duplicate names, files over 500 lines, broken relative references, and bundled script syntax.
Compare user skills with built-in and plugin skills before recommending the active global set.
Apply non-breaking fixes first. Ask before renaming skills or removing installed entries because explicit $skill-name invocations can break.
Re-run validation and summarize active, available, invalid, overlapping, and deferred items.

Codex Locations

User: ~/.agents/skills/<skill>/SKILL.md
Repository: .agents/skills/<skill>/SKILL.md
Admin: /etc/codex/skills/<skill>/SKILL.md
System and plugin skills: managed by Codex; do not overwrite them from this repository.

Codex follows symlinked skill folders. Link selected skill directories individually instead of replacing ~/.codex/skills, which may contain managed system skills.

Validation

Locate the built-in creator in the active Codex home, then run its validator:

python3 ~/.codex/skills/.system/skill-creator/scripts/quick_validate.py path/to/skill

Also validate bundled scripts without executing external actions:

bash -n path/to/script.sh
ruby -c path/to/script.rb
node --check path/to/script.js

Quality Checklist

Folder and frontmatter name are intentionally aligned.
description states what the skill does, when it triggers, and meaningful exclusions.
Portable frontmatter uses name and description; host-specific fields are added only when the current validator supports them.
SKILL.md stays under 500 lines and moves details into one-level references/ files.
Repeated or fragile operations use tested scripts.
agents/openai.yaml is added when UI metadata, MCP dependencies, or allow_implicit_invocation: false is useful.
README/catalog counts are derived from the repository rather than maintained by guesswork.

Use references/SKILL_TEMPLATE.md for the portable template. Use $skill-creator to initialize new skills or perform a substantial redesign.

이 저장소의 다른 Skills

같은 저장소

rpf

jiunbae/agent-skills

Runs an explicit pointer-document-driven review, plan, work, and feedback loop with multiple agents, scheduling independent review, verification, and implementation work concurrently and pipelining revision-fenced read-only preparation between cycles until the living plan or policy converges or 128 cycles are reached. Safe for concurrent runs from different tools against one pointer. Use only when the user explicitly invokes `$rpf`; never trigger implicitly for ordinary review, planning, implementation, commit, or deployment requests.

2026-07-3112

air-workbench

jiunbae/agent-skills

Open AIR Workbench to discover local Agent Skills, visualize and round-trip a Skill workflow as AIR (Agent Intermediate Representation), review a native Codex or Claude plan, inspect observable execution evidence, or promote a reviewed plan or trace into a Skill draft. Use only when the user explicitly asks for AIR Workbench, the legacy Workflow Studio, Skill-to-graph or graph-to-Skill conversion, visual workflow editing, plan approval, observable tracing, or plan/trace promotion. Do not trigger for ordinary Skill execution, general planning, or routine Codex/Claude requests.

2026-07-3012

korean-editor

jiunbae/agent-skills

기존 한국어 초안을 의미·사실·수치·인용·문서 구조와 작성자의 목소리를 보존하며 자연스럽게 퇴고한다. 사용자가 한국어 글의 윤문, 퇴고, 교정, 번역투 완화, 기계적이거나 AI 같은 문체 완화, 간결화, 톤 정리를 명시적으로 요청할 때 사용한다. 빈 문서에서 새 글을 쓰는 요청, 번역·요약·사실 확장, 코드 수정에는 사용하지 않는다. PR·Linear·Slack·Notion 문서는 해당 문서의 내용을 작성하는 일반 요청이 아니라 문체 편집을 명시적으로 요청한 경우에만 사용한다.

2026-07-2912

managing-vault-secrets

jiunbae/agent-skills

Retrieves and stores credentials in the user's self-hosted Vaultwarden via the `bw` CLI. Use whenever a task needs a credential the environment does not already provide — npm/PyPI publish tokens, API keys, DB passwords, registry logins — and before ever telling the user a credential is missing. Triggers on "vault 조회", "API 키 가져와", "비밀번호 저장", "secret 등록", "토큰 어디", "npm token", "publish token", "배포 토큰", "credentials", "401 unauthorized".

2026-07-2812

hello-agent

jiunbae/agent-skills

Synthetic AIR example with no private session content

2026-07-2812

background-implementer

jiunbae/agent-skills

Run bounded parallel implementation with isolated workers (native worktrees or Codex), respect a dependency DAG, and verify each worker's diff before integrating. Use only when the user explicitly requests background, parallel, or multi-agent implementation, such as "백그라운드 구현", "bg impl", or "병렬 구현". Do not trigger for ordinary implementation or code-writing requests.

2026-07-2312

name	managing-skills
description	Audits and maintains a local skill collection, including discovery, Codex compatibility, trigger scope, installation paths, bundled resources, and validation. Use for "스킬 분석", "스킬 현황", "스킬 점검", "스킬 개선", or skill ecosystem maintenance. Use the built-in skill creator for new skills and substantial rewrites.

Skill Manager

Maintain a skill repository without loading or installing every skill by default.

Workflow

Identify the source-of-truth repository and synchronize it only when the working tree is clean or the update can be proven safe.
Inventory SKILL.md files, installed links, and the skills exposed in the current session. Treat those as separate sets.
Run the current Codex validator for every skill directory.
Review descriptions for precise positive triggers and important non-trigger boundaries. Broad descriptions cause unwanted implicit invocation.
Check folder/name consistency, duplicate names, files over 500 lines, broken relative references, and bundled script syntax.
Compare user skills with built-in and plugin skills before recommending the active global set.
Apply non-breaking fixes first. Ask before renaming skills or removing installed entries because explicit $skill-name invocations can break.
Re-run validation and summarize active, available, invalid, overlapping, and deferred items.

Codex Locations

User: ~/.agents/skills/<skill>/SKILL.md
Repository: .agents/skills/<skill>/SKILL.md
Admin: /etc/codex/skills/<skill>/SKILL.md
System and plugin skills: managed by Codex; do not overwrite them from this repository.

Codex follows symlinked skill folders. Link selected skill directories individually instead of replacing ~/.codex/skills, which may contain managed system skills.

Validation

Locate the built-in creator in the active Codex home, then run its validator:

python3 ~/.codex/skills/.system/skill-creator/scripts/quick_validate.py path/to/skill

Also validate bundled scripts without executing external actions:

bash -n path/to/script.sh
ruby -c path/to/script.rb
node --check path/to/script.js

Quality Checklist

Folder and frontmatter name are intentionally aligned.
description states what the skill does, when it triggers, and meaningful exclusions.
Portable frontmatter uses name and description; host-specific fields are added only when the current validator supports them.
SKILL.md stays under 500 lines and moves details into one-level references/ files.
Repeated or fragile operations use tested scripts.
agents/openai.yaml is added when UI metadata, MCP dependencies, or allow_implicit_invocation: false is useful.
README/catalog counts are derived from the repository rather than maintained by guesswork.

Use references/SKILL_TEMPLATE.md for the portable template. Use $skill-creator to initialize new skills or perform a substantial redesign.