Przejdź do treści
Biblioteka skilli
AI i agenci · Nauka i wyjaśnianie

skill-creation-guide

/skill-creation-guide

Prowadzi przez pisanie nowego skilla: kiedy warto go tworzyć, format SKILL.md, progresywne ujawnianie i typowe pułapki.

Repozytorium
claude-plugins ↗
Licencja
MIT
Ostatnia zmiana repozytorium
29.09.2026

Ocena i bezpieczeństwo

100/ 100Klasa: dobry
Format25 / 25
Frontmatter z nazwą i opisem, poprawna nazwa, opis do 1024 znaków, nazwa zgodna z folderem.
Odkrywalność25 / 25
Opis mówi, kiedy użyć skilla, ma sensowną długość i jest w 3. osobie.
Treść35 / 35
Co najmniej 150 słów i do 500 linii, nagłówki, listy kroków, przykłady albo kod.
Higiena15 / 15
Bez zaślepek typu TODO, bez ścieżek Windows, z licencją.
Ryzykobrak sygnałów w skanie
Pliki w paczce
1
Skrypty
0
Komentarze HTML
0
Adresy URL w SKILL.md
0
Przestarzałość

Brak sygnałów. Repozytorium zmienione 29.09.2026, bez odwołań do wycofanych modeli i API.

Licencja

MIT, otwarta: można używać i zmieniać, z zachowaniem informacji o autorze.

Co robi i kiedy

Prowadzi przez pisanie nowego skilla: kiedy warto go tworzyć, format SKILL.md, progresywne ujawnianie i typowe pułapki.

Wyzwalacze z opisu

create a skillwrite a skillmake a skillnew skillSKILL.md formathow do skills workskill best practicesconvert this workflow into a skill
Oryginalny opis (EN)

Guide the user through authoring a new agent skill — when a skill is worth creating, the SKILL.md format, progressive disclosure, and common pitfalls. Also covers discovering skill opportunities from the user's work patterns and generating a SKILL.md from a description. Trigger phrases include "create a skill", "write a skill", "make a skill", "new skill", "SKILL.md format", "how do skills work", "skill best practices", "convert this workflow into a skill", "what should I automate", and "find skill opportunities". Don't use it for invoking existing skills — only when authoring a new one or und

Jak działa

  1. BE SKEPTICAL
  2. Workflow A — Discover opportunities
  3. Workflow B — Generate a skill
  4. SKILL.md structure
  5. Best practices
  6. Where to save
  7. Related skills
Sekcje z nagłówków SKILL.md, w oryginale · 116 linii, 666 słów
Podgląd SKILL.md
---
name: skill-creation-guide
description: Guide the user through authoring a new agent skill — when a skill is worth creating, the SKILL.md format, progressive disclosure, and common pitfalls. Also covers discovering skill opportunities from the user's work patterns and generating a SKILL.md from a description. Trigger phrases include "create a skill", "write a skill", "make a skill", "new skill", "SKILL.md format", "how do skills work", "skill best practices", "convert this workflow into a skill", "what should I automate", and "find skill opportunities". Don't use it for invoking existing skills — only when authoring a new one or understanding the skill format.
---

# Skill Creation Guide

Help the user author effective agent skills — single `SKILL.md` files (in the
open Agent Skills format) that auto-trigger by context and provide specialized
guidance. The format is shared across hosts (Claude Code, Cursor, and others);
only the install location differs.

This skill covers three things: deciding whether a skill is worth creating,
discovering candidates from the user's work, and generating a well-formed skill.

## BE SKEPTICAL

Not every idea makes a good skill. Evaluate before creating:

**Recurrence** — Will this be used repeatedly?
- ✅ Weekly or more → create
- ⚠️ Monthly → reconsider, may not be worth it
- ❌ One-time → skip

**Automation** — Can it actually be automated?
- ✅ Clear, repeatable process → create
- ⚠️ Needs significant judgment each time → reconsider
- ❌ Too context-dependent, each case unique → skip

**Value** — Is the cumulative time saved significant?
- ✅ Meaningful savings per use → create
- ❌ Slower to invoke than doing manually → skip

**Duplication** — Does it already exist?
- ✅ Novel, unaddressed need → create
- ⚠️ Similar to an existing skill → enhance that instead
- ❌ Already covered → skip

Don't create skills for one-time events, tasks that differ fundamentally each
time, sub-30-second manual tasks, or things the user will forget exist.

## Workflow A — Discover opportunities

When the user asks what they should automate, analyze their work patterns first.

Requires the Glean MCP tools. If they aren't visible, ask the user to set up
Glean by prompting the harness (for example, `Set up Glean for me`).

1. Gather context with Glean: `memory` (roles, responsibilities, active
   projects), `user_activity` (past ~2 weeks), and `search` for process docs
   (`"runbook OR checklist OR process owner:me"`).
2. Spawn the `work-pattern-analyzer` agent to identify repeated queries,
   frequent contexts, workflow sequences, and manual processes — and to vet each
   candidate against the recurrence/value tests above.
3. Present vetted recommendations grouped by cumulative value, and show the
   rejected candidates with reasons. A few high-quality candidates beat many
   weak ones — "no automatable patterns" is a valid outcome.

See `using-glean/reference/` for the param shape of `memory`, `user_activity`,
and `search`.

## Workflow B — Generate a skill

When the user wants to create a specific skill:

1. Clarify (only what isn't already clear): purpose, trigger conditions, tools
   it uses, and output format.
2. Spawn the `skill-generator` agent with the concept, requirements, and
   available-tools context. It returns a complete `SKILL.md` — or declines, with
   reasons, if the idea fails the skepticism tests.
3. Offer to save it to the user's host skills directory, or display it for
   review. Confirm the path and the trigger phrases that will activate it.

## SKILL.md structure

```yaml
---
name: skill-name-in-kebab-case
description: When this skill triggers — specific phrases, contexts, and use cases. Fold trigger phrases into this field; keep it thorough but under ~200 words.
---
```

```markdown
# Skill Title

Brief overview of what this skill does.

## When This Applies
- Condition / trigger phrases

## Main Content
[The workflow, guidance, or instructions — imperative form]

## Output Format (optional)
[Template for what the skill produces]
```

## Best practices

1. **Specific triggers.** "Use when reviewing pull requests, or when the user
   says 'review this PR' / 'check my code'" beats "use for code review."
2. **Progressive disclosure.** Essential action first; detail and edge cases
   below. Push deep, load-on-demand material into `reference/*.md`.
3. **Actionable, imperative content.** "Search for X", not "Searches for X."
4. **Name the tools** the skill uses (e.g. Glean `search`, `memory`; or `Grep`,
   `Read`).
5. **Stay host-agnostic.** Don't hard-code one host's slash-command syntax or
   install paths; describe the capability and let each host resolve it.

## Where to save

Skills live in the host's skills directory — e.g. `~/.claude/skills/<name>/`
(personal) or `.claude/skills/<name>/` (project) for Claude Code, the equivalent
location for other hosts, or `skills/<name>/` when authoring for a plugin
library like this repo.

## Related skills

- `work-pattern-analyzer` / `skill-generator` agents back the two workflows above.

Wersje

RepozytoriumWynikLicencja
Ta strona · gleanwork/claude-plugins100MIT
patriotai/ai-lab100MIT
gleanwork/agent-plugins100MIT

Identyczna treść występuje jeszcze w 1 innych folderach na GitHubie. Tu pokazujemy jedną, najlepiej ocenioną kopię.

Migawka bazy z 13.11.2026 · wersja paczki ab5a3c413e2f (skrót SHA-256 całej paczki)

Pobrania w ostatnich 30 dniach: 0 · ranking

Zgłoś problem z tym skillem: szkodliwa treść, błędny opis albo prośba autora o usunięcie.

Podobne skille

  • 100create-skill
    Pomaga tworzyć i pisać skille agenta Cursor oraz wyjaśnia strukturę skilla, format SKILL.md i dobre praktyki.
  • 100claude-skills
    Prowadzi przez tworzenie Agent Skills z progresywnym ujawnianiem i dobrymi praktykami: struktura skilla i progressive disclosure.
  • 100skill-authoring
    Pomaga napisać nowy skill dla agenta AI: zakres, opis dla niezawodnego wyzwalania, progresywne ujawnianie i różnica między skillem a promptem.
  • 90create-skill
    Tworzy lub restrukturyzuje skille agentów: frontmatter, routing opisu i stopniowe ujawnianie treści, na prośbę o napisanie nowego skilla.
  • 85Skill Development
    Prowadzi przez tworzenie nowego skilla do wtyczki Claude Code: strukturę, progresywne ujawnianie treści i dobre opisy.
  • 100skill-creator
    Tworzy skille w Markdown: frontmatter i treść SKILL.md, progresywne ujawnianie oraz procedurę create_skill z referencjami i szablonami.

Uwagi i zgłoszenia

Wiadomość trafi do Bartosza Sokolińskiego. Adres e-mail jest potrzebny tylko do odpowiedzi. Szczegóły w polityce prywatności.