A deliberately leaky demo skill used to show what skill-sanitizer catches before you publish. Do not ship this.
日本語の概要は準備中です。原文の説明を表示しています。
Package an AI skill, prompt workflow, or internal operating habit into a public-ready release with a catchy hook, safe redaction, useful examples, demo tasks, README copy, and launch messaging. Use when preparing a skill for GitHub, social sharing, marketplace submission, documentation, or community feedback.
インストールする前に、エージェントに与えられる指示の中身を確認できます。
Turn a useful internal skill into something a stranger can understand, try in 60 seconds, and want to share - without leaking how it really works inside.
The goal is not to publish everything. The goal is to ship the smallest delightful version that proves usefulness, and hold the rest back.
Ship the useful public core. It should be genuinely helpful on its own - a real win the user can feel today. Keep the internal edge out of the public package:
Done well, the public skill is good enough that people adopt it and say "this is useful" while the sensitive depth stays private. A public skill can and should be smaller than the internal one. If the useful part cannot survive without private details, build a toy version or do not publish.
Produce a package with these pieces:
name:
one_line_hook:
who_it_helps:
pain_it_solves:
what_it_does:
quick_start:
example_prompts:
demo_task:
safety_notes:
what_is_not_included:
launch_post:
Keep every piece concrete. A popular skill makes a user think "I can use this today," not "interesting architecture."
Split the idea into three layers:
Ship the public layer. Generalize the semi-private. Drop the private.
Prefer names that are short, action-oriented, easy to say, and specific enough to remember.
Good patterns:
bug-repro-packagerreadme-rescuedemo-script-makerrelease-note-polisherprompt-to-checklistAvoid names that depend on private project lore, internal codes, or anything only an insider would recognize.
For [audience] who struggle with [pain], this skill [action] so they can [visible outcome].
One sentence. If it needs two, the scope is too wide - narrow it.
A public skill needs a 60-second demo that:
If the demo needs setup, make a smaller demo.
3 to 5 example prompts, practical not abstract:
Use $skill-name to turn this git log into release notes.
Use $skill-name on this stack trace to make a repro checklist.
Use $skill-name to rewrite this README intro for newcomers.
Before publishing, scan the package and remove or replace:
Replace each with a toy example or a generic placeholder. Publishing is irreversible - once it is out, it can be cached, forked, and indexed even after deletion. When in doubt, leave it out.
I made [skill].
It helps [audience] do [job] without [pain].
Try it when you need to:
- [use case 1]
- [use case 2]
- [use case 3]
Example:
[short input -> short output]
Link:
Useful and light. Do not oversell.
Internal skill: a release-note generator with tuned commit-classification heuristics and team-specific tone rules.
Public version:
name: release-note-polisher
one_line_hook: For makers who ship often, this turns a messy git log into a clean, readable release note.
who_it_helps: Solo devs and small teams who dread writing changelogs.
pain_it_solves: Raw commit logs are noisy; hand-writing notes is slow and easy to skip.
what_it_does: Groups commits into Added / Fixed / Changed, drops noise, writes plain-language bullets.
quick_start: Use $release-note-polisher on your last 20 commits.
example_prompts:
- Use $release-note-polisher to turn this git log into v1.2 notes.
- Use $release-note-polisher to summarize this week's merges for a blog post.
demo_task: Paste a 15-line fake git log -> get a 3-section release note.
safety_notes: Uses only the commit text you paste; no repo access, no private data.
what_is_not_included: The internal commit-importance scoring and team tone presets are not part of this public version.
launch_post: |
I made release-note-polisher.
It turns a messy git log into a clean release note in seconds.
Try it when you need to: ship notes fast, summarize a sprint, write a changelog you actually like.
Example: 15 noisy commits -> Added / Fixed / Changed, in plain English.
The what_is_not_included line is doing real work: it is honest, sets the public boundary, and signals that private tuning is not part of the release.
# Skill Name
One sentence hook.
## What It Does
- Benefit 1
- Benefit 2
- Benefit 3
## Quick Start
```text
Use $skill-name to ...
```
## Example
Input:
```text
...
```
Output:
```text
...
```
## Not Included
This public version uses generic examples and intentionally leaves out the deeper tuning, private data, and protected internals.
GO: useful with toy examples and no protected detail.
TRIM: useful, but private details still need removing or generalizing.
HOLD: value depends on private depth or on unpublished evidence.
NO: contains secrets, client data, credentials, or core IP.
When uncertain, choose TRIM or HOLD. Publishing cannot be undone.
まだレビューはありません。使ってみた感想をお寄せください。
概要と使いどころ
A deliberately leaky demo skill used to show what skill-sanitizer catches before you publish. Do not ship this.
日本語の概要は準備中です。原文の説明を表示しています。
Analyze folders of Markdown/text research notes, Obsidian vaults, paper cards, invention notes, or memory exports as a practical knowledge map. Use when asked to find center topics, noisy or low-signal notes, possible contamination, overgrown topic groups, orphan notes, cleanup actions, or a useful public knowledge-hygiene workflow.
日本語の概要は準備中です。原文の説明を表示しています。
Pre-share leak scanner for Claude Code skills (or any folder). Scans BEFORE you publish for secrets (API keys, tokens, private keys, .env values, high-entropy strings), PII (emails/phones), local machine paths (C:\Users\<name>, /home/<name>, /Users/<name>, UNC, internal hostnames), and client/proper-noun names from a LOCAL private dictionary; also flags service-specific code (jira/github/gitlab/slack/salesforce/notion/linear/ stripe/aws/google) that should be split into a dedicated service plugin. Use when the user says "sanitize this skill", "before sharing a skill", "check for leaks", "is this skill clean", "is this safe to commit/publish/ open-source", "scrub secrets", "redact before commit", or "pre-push leak check". Cross-platform, stdlib-only Python, fail-closed.
日本語の概要は準備中です。原文の説明を表示しています。
Lightweight multi-agent research orchestration using one coordinator, three domain leads, and three lightweight research agents. Use for medium-size research, market scans, competitor comparisons, prior-art style exploration, report planning, and any task that benefits from parallel domain decomposition without running a full high-cost research lab.
日本語の概要は準備中です。原文の説明を表示しています。