RonanCodes/skills/new-app
skills/new-app/SKILL.md
--- name: new-app description: Generic "start a new project" orchestrator. Walks the canonical [[stack-decision-map]] tree (audience → app shape → scale → reactivity → EU residency → owns-the-table → connectors → payments) using AskUserQuestion, lands on a leaf, dispatches to the framework-specific skill that builds it (`/ro:new-tanstack-app`, `/ro:new-astro-app`, `/ro:new-mcp-server`, ...). Use when the user wants to start, create, scaffold, bootstrap, or begin a new project but isn't sure whic
tools
Updated Jun 29, 2026
$ install --global
skillsauthnpx skillsauth add RonanCodes/ronan-skills skills/new-appInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: Jun 29, 2026, 2:35 AM200.2s1 file scanned
SKILL.md
- name:
- new-app
- description:
- Generic "start a new project" orchestrator. Walks the canonical [[stack-decision-map]] tree (audience → app shape → scale → reactivity → EU residency → owns-the-table → connectors → payments) using AskUserQuestion, lands on a leaf, dispatches to the framework-specific skill that builds it (`/ro:new-tanstack-app`, `/ro:new-astro-app`, `/ro:new-mcp-server`, ...). Use when the user wants to start, create, scaffold, bootstrap, or begin a new project but isn't sure which stack fits — or when you'd rather walk the decision tree once than memorise four sibling skills. Routes to the right downstream skill so users never need to know the catalogue.
- category:
- project-setup
- argument-hint:
- [<app-name>] [--from-doc] [--explain]
- allowed-tools:
- Bash(cat *) Bash(git *) Read AskUserQuestion
New App (decision-tree dispatcher)
A thin orchestrator that walks [[stack-decision-map]] and hands off to the framework-specific skill once the user lands on a leaf. Doesn't scaffold anything itself — every actual file write happens in the dispatched-to skill.
If you already know which framework you want, skip this and call the leaf skill directly:
/ro:new-tanstack-app— server-leaning SaaS / app/ro:new-astro-app— static-leaning marketing / portfolio / blog/ro:new-mcp-server— MCP server (McpAgent on Workers, optional--with-card)
This skill is the "I'm not sure" entry point.
Every dispatch path scaffolds a /styleguide route by default (role-gated to superadmin + staff when auth is wired, dev-only otherwise). Downstream skills like /ro:new-tanstack-app and /ro:new-astro-app invoke /ro:design-system-create --showcase as part of their flow; this orchestrator doesn't need to ask about it.
Usage
/ro:new-app # walk every question
/ro:new-app my-site # pre-fill the app name
/ro:new-app --from-doc # re-read stack-decision-map.md before walking (don't trust cached snapshot)
/ro:new-app --explain # at each fork, print the rationale from the canon table
Canonical source
The decision tree this skill walks lives in [[stack-decision-map]] (LLM wiki research vault). Treat that doc as the source of truth — the prompts below reflect a 2026-05-05 snapshot. If the wiki tree changes (new framework leaf, removed alt, etc.), update the dispatch matrix at the bottom of this file rather than re-deriving the questions from scratch.
If --from-doc is passed, resolve the wiki repo with bash "${CLAUDE_PLUGIN_ROOT:-$HOME/Dev/ronan-skills}/scripts/resolve-repo.sh" LLM_WIKI (or use $LLM_WIKI_DIR) and read the source mermaid block under $LLM_WIKI_DIR/vaults/llm-wiki-research before asking anything — that's the authoritative version.
Process
1. Walk the decision tree
Use AskUserQuestion for each fork. Keep the wording faithful to the wiki tree:
Q1 — Audience (Q_AUDIENCE):
- Power user with skill kit (default; this skill exists for you)
- First-app vibe coder (escape-hatch — recommend the Alex Finn stack, see [[stack-canon-vs-alex-finn]], skip the rest of the tree)
Q2 — App shape (Q_SHAPE):
- Web SaaS (server-leaning, has dynamic data, auth, etc.)
- Marketing / landing / portfolio / blog (static-leaning, content-heavy)
- MCP server (LLM tool/resource provider)
- 2D game
- 3D experience
- Mobile-first PWA
If Marketing/static, ask Q3a — Existing site?
- No, fresh build → dispatch
/ro:new-astro-app <app-name> - Yes, existing live site → dispatch
/ro:migrate-to-astro
If Web SaaS, walk Q_SCALE → Q_REACTIVE → Q_EU → Q_OWNTABLE → Q_CONNECTORS → Q_PAYMENTS — but don't re-implement those questions here. Just dispatch /ro:new-tanstack-app <app-name> and let it walk its own --interactive flow with the same questions. (Different sub-skill, same questions, single source of truth.)
If MCP server: dispatch /ro:new-mcp-server <app-name>.
If 2D game | 3D experience: those leaves point to tbd:phaser-game / tbd:r3f etc. — print the leaf name and a short note that the dedicated skill is pending.
If Mobile-first PWA: dispatch /ro:new-tanstack-app <app-name> --pwa (which then composes /ro:pwa-install post-bootstrap).
2. Confirm the dispatch
Before invoking the chosen skill, summarise the decisions in one paragraph and print the exact downstream invocation. The user can interrupt if anything looks wrong.
Based on your answers:
- audience: power-user
- shape: marketing site
- migrate: no, fresh build
About to run: /ro:new-astro-app <app-name>
Continue? (yes/edit-decisions/no)
3. Hand off
Invoke the dispatched skill. From here on, this skill is silent — the leaf skill drives.
Dispatch matrix (current as of 2026-05-05)
| User picks | Then... | Dispatches to |
|---|---|---|
| Audience: vibe-coder | (no further questions) | (recommend Alex Finn stack, no skill dispatch) |
| Shape: Web SaaS | Walk SaaS-specific tree inside the skill | /ro:new-tanstack-app <app-name> --interactive |
| Shape: Marketing → fresh | (none) | /ro:new-astro-app <app-name> |
| Shape: Marketing → migrate | (none — migrate-app handles the source-side qs) | /ro:migrate-to-astro |
| Shape: MCP server | (sub-skill picks transport) | /ro:new-mcp-server <app-name> |
| Shape: Mobile PWA | (sub-skill composes PWA install) | /ro:new-tanstack-app <app-name> --pwa |
| Shape: 2D game | (no skill yet) | placeholder: tbd:phaser-game |
| Shape: 3D experience | (no skill yet) | placeholder: tbd:r3f / tbd:babylon |
When the wiki gains a new framework leaf, add a row above and a question/answer pair in step 1.
See also
[[stack-decision-map]]— the canonical tree this skill walks/ro:migrate-app— the migration-side sibling (same dispatcher pattern, different leaf set)/ro:new-tanstack-app,/ro:new-astro-app,/ro:new-mcp-server— the leaf skills this dispatcher routes to/ro:setup-logging— every scaffolded app should run this for the diagnosable-by-default observability baseline (structured logging + trace_id + CF observability); pairs with/ro:diagnosecanon/auth-guards.md(MANDATORY with auth) — every login-gated page needs a server-side route guard; signed-out users must be redirected, never shown a gated page that then 401/500s
Related Skills
RonanCodes/skills/linear-pipeline
testing
VerifiedTrustedCommunity
--- name: linear-pipeline description: The Fable orchestrator for a single dispatched Linear ticket. Holds almost no context itself; it receives `--issue <ID> --detached`, decides the stage sequence, and fans out a sub-agent per stage, passing forward only each stage's artifact (never re-derived, never inlined into its own context). Step zero, before any planning or stage routing, is a boundary triage against `canon/security-boundary.md` (#199): a match tags Ronan Connolly and stops the run, no
SKILL.mdUpdated Jul 5, 2026
RonanCodes/skills/in-your-face
development
VerifiedTrustedCommunity
--- name: in-your-face description: Capture a chat-only answer into a durable artifact (markdown + HTML, PDF when cheap) and launch it automatically so the user cannot miss it. Use when user says "in your face", "don't let me lose this", "save that answer", "make that durable", or right after answering a substantive side question (a recipe, comparison, how-to, or generated prompt) that would otherwise die with the context. category: workflow argument-hint: [--no-open] [--vault <short>] [hint of
SKILL.mdUpdated Jul 5, 2026
RonanCodes/codex
tools
VerifiedTrustedCommunity
One-shot headless OpenAI Codex CLI calls for background/admin AI tasks — summaries, classification, extraction, admin glue. The default engine for anything that runs AI constantly in the background (daemon-driven, per-event), because it bills the flat ChatGPT subscription instead of Claude usage or per-token API spend, and it keeps working while Claude is rate-limited. NEVER for coding — coding stays Claude. Use when a skill or daemon needs a cheap always-on AI call, when the user says "use codex", "ask codex", "codex as backup", or when building a background summarizer/classifier into a listener or loop. Reads auth from ~/.codex/auth.json (ChatGPT account, no API key).
SKILL.mdUpdated Jul 5, 2026
RonanCodes/warranty-check
research
VerifiedTrustedCommunity
Turn a warranty rejection, repair quote, or RMA email into a cited decision brief — legal read (NL/EU consumer law), is the part user-serviceable, live part and new-unit prices, repair-vs-DIY-vs-new economics, before-you-send-it checklist, deadlines. Use when the user pastes or screenshots a repair quote, warranty rejection, "not covered" email, onderzoekskosten fee, or asks "should I repair or replace this".
SKILL.mdUpdated Jul 4, 2026