Install
$ agentstack add skill-shivae372-claude-bootstrap-tips ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ● Environment & secrets Used
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Purpose
Teach the user 3–5 high-impact habits, chosen to fit their tech level and what this project actually has. Tips live in this file (no external tips.json needed).
When To Use This
- The user runs
/tipsor asks "how do I get better at this / what am I doing wrong?" analyze-repocalls it at the end of a scan (show only 2–3, the most relevant)
Steps
- If
USER_PROFILE.jsonexists, readtech_levelandprimary_goalsto bias selection;
otherwise assume tech_level 3.
- Look at the project: are there tests? agents? a long CLAUDE.md? a SESSION_STATE.md?
- Pick 3–5 tips from the catalog below, preferring ones the user can act on right now.
- Present them scannably: a bold category, the tip, and a one-line
→ example.
Tip Catalog
Tokens / context (tech 1–5)
- 💡 Delegate exploration to the
explorersub-agent — its context is isolated, so reading a
big codebase doesn't bloat your main session. → "explore where auth is handled"
- 💡 Ask for targeted reads, not whole files. → "show lines 40–80 of server.ts", not "open server.ts"
- 💡
/compact(or let the PreCompact hook checkpoint) before starting a new sub-task.
Workflow (tech 2–5)
- 💡
/planbefore building anything non-trivial — agree on scope first. → "/plan add OAuth login" - 💡
/shipbefore committing — it gates tests + review + security in one pass. - 💡 Keep
SESSION_STATE.mdcurrent so a fresh session resumes instantly (the SessionStart hook reads it).
Quality (tech 3–5)
- 💡
/reviewyour diff before opening a PR; fix 🔴/🟠 findings first. - 💡 Put hard rules in hooks, not just CLAUDE.md — hooks are 100% reliable, prose is ~70%.
Prompting (tech 1–4)
- 💡 State the goal AND the constraints. → "add caching, but don't change the public API"
- 💡 When output is wrong, say what you expected — don't just say "no".
Safety (tech 1–5)
- 💡 The safety-check + secret-detector hooks have your back, but still review destructive commands.
- 💡 Never paste real secrets into chat — reference
process.env.X/os.environ["X"]instead.
Notes
- From
analyze-repo, show only 2–3 and label them "Quick wins for this setup". - Don't repeat the same tips every time; vary by category.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: shivae372
- Source: shivae372/claude-bootstrap
- License: Apache-2.0
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.