Przejdź do treści
Biblioteka skilli
Programowanie · Automatyzacja

get-started

/get-started

Prowadzi przez konfigurację hackathonu od czystego klonu po działającą aplikację na GitOps, pokazując plan dnia i pierwsze kroki.

Repozytorium
roloe ↗
Licencja
Apache-2.0
Ostatnia zmiana repozytorium
10.11.2026

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
5
Przestarzałość

Brak sygnałów. Repozytorium zmienione 10.11.2026, bez odwołań do wycofanych modeli i API.

Licencja

Apache-2.0, otwarta: można używać i zmieniać, z zachowaniem licencji i informacji o zmianach.

Co robi i kiedy

Prowadzi przez konfigurację hackathonu od czystego klonu po działającą aplikację na GitOps, pokazując plan dnia i pierwsze kroki.

Oryginalny opis (EN)

Get this person set up for the hackathon, from a fresh clone to their app live through GitOps, and lay out the day, the way a coach at the Radar table would. Use when they ask to get set up, check their setup, or what to do first, when they ask what they're looking at in Radar, and at the start of a first session in this repo.

Jak działa

  1. Check (read only)
  2. Set up (ask once, then run it through)
  3. The day
Sekcje z nagłówków SKILL.md, w oryginale · 42 linii, 658 słów
Podgląd SKILL.md
---
name: get-started
description: Get this person set up for the hackathon, from a fresh clone to their app live through GitOps, and lay out the day, the way a coach at the Radar table would. Use when they ask to get set up, check their setup, or what to do first, when they ask what they're looking at in Radar, and at the start of a first session in this repo.
---

# Get set up for the hackathon

Check first, then set up. Keep replies short. At a check that fails, say in a sentence or two what to do, and wait.

## Check (read only)

1. **Team.** This repo is already set up for team `roloe` (namespace `team-roloe`); there is no setup.sh to run.

2. **Radar connected?** If you have no Radar tools, or a call says the server needs authentication, the person signs in; you can't do it for them.
   - Claude Code: if `/mcp` doesn't list radar-cloud, run `claude mcp add --scope user radar-cloud --transport http https://api.radarhq.io/mcp` in a terminal and restart Claude Code. Then `/mcp` > radar-cloud > **Authenticate**.
   - Codex: `codex mcp add radar-cloud --url https://api.radarhq.io/mcp`, then restart Codex.
   - Other agents: in Radar Cloud, open **Agents** and pick theirs.

   In the browser they sign in with the email their team listed, and under **Organization access** choose **Cloud Native Hackathon**.

3. **Cluster found?** Call `list_clusters` and take the entry whose `name` is exactly `<team>`; its `id` is the `cluster` for every other call. If it's missing and so is `cnh`, the sign-in went to another organization: sign in again (`/mcp` > radar-cloud) and choose Cloud Native Hackathon. If it isn't offered, they haven't accepted the invite email yet. If `cnh` is there but `<team>` isn't, check the spelling and `https://join.cnh.radarhq.io/status/<team>`.

4. **Readable?** Call `list_resources` with `kind: deployments`, `namespace: team-<team>`. If Radar refuses or can't read any namespace right after sign-up, that's normal: access lands within a minute of signing up, and Radar remembers a refusal for 2 minutes. Say so, wait 2 minutes, retry, for up to about 5 minutes. After that: check they signed in with an email the team listed, then send them to the Radar table. In Radar Cloud the same lag shows as **No Namespace Access … contact your administrator**: wait two minutes and reload.

5. **The first screen.** Their cluster is the card named exactly `<team>` on Radar Cloud's Home; the rest are other teams' and the platform's (`cnh`). Call `issues` and sort it against "What every team sees" in AGENTS.md: name the known noise in a line, and point out anything else.

## Set up (ask once, then run it through)

Say what's left of the list below and ask once: "Shall I set it up? I'll push to your repo and deploy through Radar." Skip what's already done (the Application `app` in `team-<team>`, `get_resource` group `argoproj.io`; `STORY.md`). When they say go, run to the end without stopping:

1. The repo is already set up for team `roloe`; skip straight to the next step.
2. Follow the `ship` skill from step 0: it waits for the build and its `deploy:` commit, applies `argocd/application.yaml`, syncs, waits for the new pods (which proves the image is public), and checks that the URL shows the pushed commit.
3. Record the "before": `get_cluster_audit` with `namespace: team-<team>`. Write `STORY.md` with a heading `## Before` (local date and time; the link `https://app.radarhq.io/checks?scope=<cluster-id>&namespace=team-<team>`; findings by severity; one line per finding: check and resource), commit and push it. Markdown alone doesn't trigger a build.

Report the URL and the version it shows.

## The day

In a few lines:
- **Before:** screenshot Radar Cloud > the cluster > **Checks**. The findings on the platform we installed are real Radar findings, worth a look, but not theirs to fix. Their own before and after is **Namespace** `team-<team>`. Link: `https://app.radarhq.io/checks?scope=<cluster-id>&namespace=team-<team>`.
- **Build** what they came to build: README, "What Radar can do for you". You ship each change with the `ship` skill.
- **One AI investigation** (the judges look for it; an all-clear on a healthy app counts). Give the resource's Radar Cloud link (AGENTS.md has the format): the sparkle button at the top right, next to Refresh, starts it. It only says **Investigate** when the resource has a problem.
- **By 5:30:** ask you for the story (the `demo-story` skill).

Then ask what they want to build.

Wersje

RepozytoriumWynikLicencja
Ta strona · WestonBDev/roloe100Apache-2.0
firecrawl/firecrawl-cursor-plugin100MIT
hibira-io/agent-plugin100MIT
keboola/ai-kit100MIT
remotivated/job-hunt-skills100MIT
aeonfun/aeon100MIT
coupler-io/skills100MIT
theopenapps/openapps-mcp100MIT
motleyai/motley-skills100MIT

Migawka bazy z 13.11.2026 · wersja paczki 66c6b9dc336b (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

  • 100local-setup-author
    Tworzy, sprawdza i aktualizuje przewodnik LOCAL_SETUP.md od świeżego klonu do działającej aplikacji i pętli edycji.
  • 100matlab-design-radar
    Uruchamia i steruje aplikacją Radar Designer w MATLAB-ie przez MCP: projektowanie systemów radarowych, konfiguracja parametrów, porównanie typów radarów i analiza wskaźników wydajności.
  • 100foundryvtt-module
    Prowadzi moduł FoundryVTT od pomysłu do repozytorium przyjętego w gitops: szkielet, utworzenie i zasilenie repo oraz otwarcie PR adopcji gitops.
  • 100radar-setup
    Instaluje Project Radar na koncie Claude: czyta lub buduje konfigurację, publikuje panel i tworzy dzienne zadanie z Linear.
  • 100b2c-onboarding
    Prowadzi przez start z Salesforce B2C Commerce: instalacja B2C CLI, konfiguracja środowiska deweloperskiego, połączenie sandboxa i wdrożenie pierwszego cartridge.

Uwagi i zgłoszenia

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