settings — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited settings (Hook) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
A Claude Code starter template. Fork it, run /onboarding, and Claude Code will configure itself for your project — interview you about what you're building, write a tailored CLAUDE.md, set sensible permissions, and hand off cleanly.
/onboardingThe first run takes around 5–10 minutes — most of that is a friendly discovery conversation about your project, how you work, and what you'd like Claude to help with most. Re-run any time to refresh.
After onboarding finishes, close and reopen Claude Code once so the new hooks take effect.
You don't need to memorize slash commands. Just say what you want in plain English. The slash commands in parens are shortcuts if you prefer.
| Phrase | Command | What it does |
|---|---|---|
| "What can you do?" | /help-me | Reminds you what commands you have |
| "Remember to..." | /remember X | Saves a preference across sessions |
| "Forget about..." | /forget | Removes a saved preference |
| "Wrap up" | /wrap | Recaps the session for next time |
| "Update my setup" | /update | Checks your setup against current best practices |
| "Save a checkpoint" (opt-in) | /checkpoint | Saves a recoverable snapshot before risky changes |
| "Undo that" (opt-in) | /undo | Reverses the last change |
The two opt-in commands appear only if you say yes to the safety net during onboarding (requires git).
Kickstart writes a short recap at the end of each working session — automatically when you close cleanly, or by saying "wrap up". The next time you open Claude Code, you'll see a brief "Where we left off" note so you can pick up without re-explaining context.
If you tell Claude "remember to..." or "from now on, always...", the preference goes into CLAUDE.md and survives across sessions.
Run this once to enable weekly audits:
/loop weekly /update/update runs two checks in parallel:
.claude/knowledge.md (the baseline).You get a plain-language summary. Nothing changes without your approval.
.claude/skills/ — slash-command skills (the ones in the table above).claude/agents/ — the curator and caretaker subagents /update calls.claude/skill-templates/ — the safety-net skills, staged until onboarding activates them.claude/settings.json — permissions and the SessionStart / Stop hooks that drive the recap patternYou don't need to know any of this to use Kickstart — but it's all here if you want to peek under the hood.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.