SVK-setup
Skill19 starsChanged 9 months agoMITView on GitHub
SVK Setup: Guided onboarding skill that interviews users about their experience and project scope, recommends tools from a tiered catalog (MCPs, plugins, skills), walks through installation of third-party tools and SVK core skills (SOS, GL, DB, BOK), and generates a personalized reference document. The first thing a new SVK user runs. Run /SVK-setup for the full guided flow, or /SVK-setup:interview to begin.
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 4 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 25
Look for .claude/svk-meta.json in the current project.
- Very specific: line 26
Read the svk_repo path and installed_version from the metadata file.
- Very specific: line 28
Run: git -C <svk_repo> tag --sort=-v:refname | head -1 to get the latest tag.
- Distinctive: line 30
If they match, skip — the user is up to date.
- Distinctive: line 31
If the latest tag is newer, show this message ONCE (never repeat in this session):
- Distinctive: line 37
If the git commands fail (offline, repo moved, etc.), skip silently.
- Distinctive: line 39
Important: Do NOT block or delay the user's actual command.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
<!-- SVK Version Check — runs once per session on first skill invocation --><svk-version-check>
**On first invocation this session**, silently perform a version check:- Look for
.claude/svk-meta.jsonin the current project. If it doesn't exist, skip this check entirely. - Read the
svk_repopath andinstalled_versionfrom the metadata file. - Run:
git -C <svk_repo> fetch --tags --quiet 2>/dev/null - Run:
git -C <svk_repo> tag --sort=-v:refname | head -1to get the latest tag. - Compare the installed version against the latest tag (strip the leading
v). - If they match, skip — the user is up to date.
- If the latest tag is newer, show this message ONCE (never repeat in this session):
SVK Update Available: v{latest} is available (you're on v{installed}).
- Update now: I can pull and reinstall the changed skills in this session
- Update later: Start a new chat and run
/SVK:update
- If the git commands fail (offline, repo moved, etc.), skip silently. Never show errors from version checking.
Important: Do NOT block or delay the user's actual command. Perform this check, show the notification if needed, then proceed with the command they invoked.
</svk-version-check>
SVK Setup
A guided onboarding skill that sets up your entire Solana development environment — not just SVK, but the full ecosystem of plugins, MCPs, and skills needed for productive Solana development.
"The first 30 minutes determine whether someone stays or bounces. Make them count."
Getting Started
SVK Setup runs as a 4-phase guided flow. The full command runs all phases in sequence; sub-commands let you re-run individual phases.
Quick Start
/SVK-setupThis begins with a short interview about your experience and project, then recommends tools, walks through installation, and generates a reference doc. Follow the prompts.
Full Pipeline
┌─────────────────────────────────────────────────────────────────────┐
│ SVK SETUP v1.0 │
├─────────────────────────────────────────────────────────────────────┤
│ │
│ /SVK-setup:interview Phase 1 — User Interview │
│ ═══════════════════ 5-7 questions, one at a time │
│ Experience gauging: git, Solana, Claude Code │
│ Project scoping: new/existing, type, frontend/backend, team │
│ Output: .svk/SETUP_PROFILE.json │
│ │ │
│ ▼ │
│ /SVK-setup:recommend Phase 2 — Tiered Recommendations │
│ ═══════════════════ Three tiers based on profile │
│ Essential: always recommended (GSD, Superpowers, Safety Net, etc) │
│ Recommended: profile-dependent (Trail of Bits, Context7, etc) │
│ Optional: shown but not pushed (Figma, Vercel, Docker, etc) │
│ Output: .svk/SETUP_RECOMMENDATIONS.json │
│ │ │
│ ▼ │
│ /SVK-setup:install Phase 3 — Installation Walkthrough │
│ ═══════════════════ One category at a time │
│ Explain → choose → install → verify per tool │
│ Category order: Workflow → Safety → Solana → Search → Memory → │
│ Security → Frontend → Backend → DevOps → Utility │
│ Output: .svk/SETUP_INSTALLED.json │
│ │ │
│ ▼ │
│ /SVK-setup:reference Phase 4 — Reference Document │
│ ═══════════════════ Personalized setup reference │
│ Getting started, installed tools, skipped tools, quick reference │
│ Output: docs/svk-setup-reference.md │
│ │
└─────────────────────────────────────────────────────────────────────┘Typical Workflow
For first-time users, just run /SVK-setup and follow the prompts. The full flow takes 10-20 minutes depending on how many tools you install.
To re-run individual phases later:
/SVK-setup:interview— Re-do the interview (resets profile)/SVK-setup:recommend— Re-generate recommendations from existing profile/SVK-setup:install— Install additional tools/SVK-setup:reference— Re-generate the reference document
Foundation Patterns
| Pattern | How SVK Setup Uses It |
| Progressive Disclosure | Tool catalog indexed by category, loaded per-category during walkthrough |
| Signal-Based Indexing | Interview answers drive which catalog entries are surfaced |
Output Structure
SVK Setup writes state to .svk/ in the project root and the reference doc to docs/:
.svk/
├── SETUP_PROFILE.json # User profile from interview
├── SETUP_RECOMMENDATIONS.json # Tiered tool recommendations
└── SETUP_INSTALLED.json # What was installed/skipped
docs/
└── svk-setup-reference.md # Personalized reference documentUser Profiles
The interview produces one of three profiles that tune all subsequent behavior:
| Profile | Git | Solana | Claude Code | Behavior |
| Beginner | None/Basic | No/Learning | New | Extra explanations, guardrails emphasized, learning resources |
| Intermediate | Comfortable | Learning/Shipped | Some | Standard explanations, full recommended tier |
| Advanced | Comfortable | Shipped | Power user | Minimal hand-holding, full catalog available, cherry-pick |
Tool Tiers
| Tier | Selection Logic |
| Essential | Always recommended regardless of profile |
| Recommended | Shown based on project type and experience |
| Optional | Listed for awareness, not pushed |
See resources/tool-catalog.md for the complete registry.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

