Skip to content
Skills Library
Programming · Learning & explanation

start

/start

Guided onboarding for a new or returning user. Use when the user just cloned or installed the studio, has no idea or stack yet, asks where to begin, or wants to bring in an existing website.

Licence
MIT
Last repository change
09.11.2026

Assessment and security

90/ 100Class: good
Format25 / 25
Frontmatter with name and description, valid name, description up to 1,024 characters, name matching the folder.
Discoverability25 / 25
Description explains when to use the skill, has a reasonable length and uses third person.
Content25 / 35
At least 150 words and at most 500 lines, headings, step lists, examples or code.
Hygiene15 / 15
No placeholders such as TODO, no Windows paths, with a licence.
Riskno signals detected in the scan
Files in this package
1
Scripts
0
HTML comments
0
URLs in SKILL.md
0
Deprecation

No signals detected. Repository changed on 09.11.2026, without references to retired models or APIs.

Licence

MIT, open: use and modify while retaining author attribution.

What it does and when to use it

Guided onboarding for a new or returning user. Use when the user just cloned or installed the studio, has no idea or stack yet, asks where to begin, or wants to bring in an existing website.

Arguments

/start [no arguments]
Original author description

Guided onboarding for a new or returning user. Use when the user just cloned or installed the studio, has no idea or stack yet, asks where to begin, or wants to bring in an existing website.

How it works

  1. Phase 0: Plugin install check
  2. Phase 1: Read the project
  3. Phase 2: Where are you starting from?
  4. Phase 3: Project type
  5. Phase 4: Rigor
  6. Phase 5: Automation
  7. Phase 6: Confirm and record
  8. Phase 7: The one next step
Original SKILL.md headings · 101 lines, 757 words
SKILL.md preview
---
name: start
description: "Guided onboarding for a new or returning user. Use when the user just cloned or installed the studio, has no idea or stack yet, asks where to begin, or wants to bring in an existing website."
argument-hint: "[no arguments]"
user-invocable: true
allowed-tools: Read, Glob, Grep, AskUserQuestion, Bash(mkdir -p .claude/scripts*), Bash(cp *), Bash(bash .claude/scripts/project-state.sh*), Bash(bash .claude/scripts/set-stage.sh *), Bash(bash .claude/scripts/config-set.sh *)
model: sonnet
---

# Guided Onboarding

The entry point. It assumes no idea, no stack and no experience. It reads the
project, asks where the user is starting from, records the stage, rigor and
automation (only after an explicit yes), and names **one** next step. It never
runs that step itself.

It writes only `project.name`, `project.type`, `project.stage` (+
`production/stage.txt`), `modes.rigor` and `modes.automation`, all through the
helper scripts. It never writes the knobs rigor expands to (see
`.claude/docs/config-resolution.md`).

Ask with the question tool. Without one, list numbered options and end the
turn. The question is the last thing in the reply. A skipped or dismissed
question is not an answer: ask again in plain text.

## Phase 0: Plugin install check

If `.claude/scripts/yaml-get.sh` does not exist, the studio was installed as a
plugin into an existing repo and the helpers are not in the project yet. This
skill's base directory is `<plugin>/.claude/skills/start/`, so the plugin's
helpers are at `<plugin>/.claude/scripts/` and its config template at
`<plugin>/project.yaml`. Say so, and ask: "May I copy the studio's helper
scripts into `.claude/scripts/` and create `project.yaml`?" On yes, `mkdir -p
.claude/scripts` and copy them (never overwrite an existing `project.yaml`).
Without them, every skill still works but cannot read or save settings.

## Phase 1: Read the project

Run `bash .claude/scripts/project-state.sh` and keep the rows; do not show them yet.

- `stack_*` set, or `manifest:` rows present → there is (or will be) code.
- `brief` / `prd` → discovery work exists.
- `origin_url` containing `claude-code-web-studios` (case-insensitive) → the
  clone still points at the template repo; a push would go there. Warn once.

**Returning user** (`stack_frontend` or `stack_backend` set **and** `prd: yes`):
ask nothing. Summarise name, type, stage, stack and rigor in three lines,
suggest `/help`, add the template-remote note if it applies, and stop.

## Phase 2: Where are you starting from?

Ask: "Welcome to Web Studios! Where are you starting from?"
- `A) No idea yet`: I want to build something but don't know what.
- `B) Rough idea`: e.g. "a booking site for dog groomers".
- `C) Clear concept`: I know the product, users and core features.
- `D) Existing website or codebase`: there's already code, docs or a live site.

If the answer contradicts Phase 1, say what you found once and re-ask (D on an
empty repo; A with `manifest:` rows on disk).

## Phase 3: Project type

Skip for D when the type is obvious from the code; otherwise ask:
"What kind of project is it?" → `SaaS / web app`, `Marketing / landing site`,
`E-commerce store`, `API / backend service` (multi-select allowed → `mixed`).
For A, offer "Not sure yet" and leave `project.type` unset.

## Phase 4: Rigor

Skip if `modes.rigor` is already set. Ask: "How much process do you want?"
- `Minimal`: prototype or side project. Checks warn but don't block; docs are brief.
- `Standard (Recommended)`: a real product. TDD for logic, gates required, hooks block secrets and broken commits.
- `Full`: production with users' money or data. Every check blocks, all six launch audits, director sign-off at gates.

## Phase 5: Automation

Skip if `modes.automation` is already set. Ask: "How often should I check with you before writing files?"
- `Ask every time (Recommended)`: you approve each file.
- `Batch`: approve a phase's changeset once.
- `Auto for docs`: orchestrated skills write the docs they name without re-asking; code and anything outward-facing still asks.
(values: `ask`, `batch`, `auto`)

## Phase 6: Confirm and record

Show a short summary (starting point, type, stage, rigor, automation) and the
stage you infer:
- A, B, C → `discovery`
- D → `discovery` if there is no PRD, `build` if there is code and a PRD,
  otherwise let `/adopt` decide (record nothing yet)

Ask "May I save these to `project.yaml`?" On yes, run `config-set.sh` for each
value and `set-stage.sh <stage>`.

## Phase 7: The one next step

| Start | Next step | What it does |
|---|---|---|
| A | `/brainstorm open` | Explores problems worth solving and writes `docs/product/brief.md` |
| B | `/brainstorm <your idea>` | Sharpens the idea into a brief |
| C | `/write-prd` | Turns the concept into a PRD (it will ask for anything missing) |
| D | `/adopt` | Detects the stack, writes `project.yaml`, maps what exists |

Then show the road ahead in one line:
**Discovery → Design (`/setup-stack`, `/design-system`, `/working-prototype` + `/test-session`, `/feature-spec`) →
Build (`/dev-story`) → Launch (`/launch-checklist`) → Grow.**

Mention `/install-deps` once (optional design + testing skills) and that
`/help` always names the next step. End there.

Versions

RepositoryScoreLicence
This page · chatquill/claude-code-websites90MIT
tilnoteio/tilnote-blog100MIT
dedujeong/gemba100Apache-2.0
nextwork-projects/ai-video-editor-cooked100MIT
wrg32786/aigent-os100MIT
T-X-R/codex-game-studio100MIT
alienfast/claude100MIT
ratomirjovanovic/pandaward100AGPL-3.0
darioabc/massimo100MIT

Database snapshot: 13.11.2026 · package version ca034bbcafad (SHA-256 digest of the entire package)

Downloads in the last 30 days: 0 · ranking

Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.

Similar skills

  • 81game-studio
    Set up Codex Game Studio for a new game or an existing game workspace, then route to the installed game development workflows. Use when the user wants to make a game or enable this toolkit in a game project.
  • 95onboarding
    Guide new users through the complete Flocks setup process via conversation — covering ThreatBook MCP/API, security tool APIs, IM channels, and scenario demos. Also runs system health inspection for returning users. Trigger when the user sends "请启动新手引导流程", "start onboarding", or…
  • 100adopt
    Brings an existing website or codebase into the studio: detects the stack, proposes project.yaml values, maps existing docs and code, and proposes the stage. Use when the user has an existing site, repo or partial project and wants to use the studio on it (brownfield).
  • 100worktree-task-start
    Use when the user asks to start a backlog task, begin work on an issue, pick up issue N, or run `make task`. Triggers on "start task", "begin issue", "pick up issue N", "work on #N".
  • 100tech-stack-research
    [Architecture] Use when a workflow step or the user asks for tech-stack research: alternatives, trade-offs and user-validated recommendations.
  • 100tech-stack-research
    [Architecture] Use when a workflow step or the user asks for tech-stack research: alternatives, trade-offs and user-validated recommendations.

Feedback and reports

Your message goes to Bartosz Sokoliński. An email address is needed only for a reply. See the privacy policy.