sebabottero/ai-setup
Make AI coding agents ask before hard-to-undo actions, and cover a 6-step cycle: decide, guardrails, build, review, human gate, learn. Claude Code · OpenCode · Codex
stack: Python
ARCHETYPE
Skill Collector
Ten-plus skills loaded on demand. A procedural-knowledge library.
Copy this rig
# review before running: this installs third-party code
$ npx degit sebabottero/ai-setup/.claude ./rig-ai-setup # inspect, then merge into .claude/ MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.
$ npx degit sebabottero/ai-setup/adapters/claude/skills/tune-setup .claude/skills/tune-setup $ npx degit sebabottero/ai-setup/core/skills/ai-audit .claude/skills/ai-audit $ npx degit sebabottero/ai-setup/core/skills/coach .claude/skills/coach $ npx degit sebabottero/ai-setup/core/skills/convert-to-markdown .claude/skills/convert-to-markdown $ npx degit sebabottero/ai-setup/core/skills/grillme .claude/skills/grillme $ npx degit sebabottero/ai-setup/core/skills/pr-review-ticket .claude/skills/pr-review-ticket $ npx degit sebabottero/ai-setup/core/skills/repo-guardrails .claude/skills/repo-guardrails $ npx degit sebabottero/ai-setup/core/skills/session-handoff .claude/skills/session-handoff $ npx degit sebabottero/ai-setup/core/skills/setup-commit .claude/skills/setup-commit $ npx degit sebabottero/ai-setup/core/skills/ticket-flow .claude/skills/ticket-flow $ npx degit sebabottero/ai-setup/core/skills/validate .claude/skills/validate
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 16.2k rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(./.env.*)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(git push --force:*)",
"Read(**/*.pem)",
"Read(.env)",
"Read(~/.aws/**)",
"Read(**/*.key)",
"Read(**/.env.*)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Bash(rm -rf *)",
"Bash(sudo *)",
"Bash(sudo:*)",
"Bash(rm -rf /)",
"Bash(git push -f:*)",
"Read(.env.*)",
"Bash(git reset --hard*)",
"Bash(rm -rf /*)"
],
"ask": [
"Bash(git push:*)",
"Bash(git push *)",
"Bash(gh pr merge *)",
"Bash(git commit:*)",
"Bash(rm *)",
"Bash(gh pr merge:*)",
"Bash(gh pr create *)",
"Bash(git reset --hard:*)",
"Bash(npm publish:*)",
"Bash(git rebase *)"
]
}
} Skills (11)
tune-setupai-auditcoachconvert-to-markdowngrillmepr-review-ticketrepo-guardrailssession-handoffsetup-committicket-flowvalidate
Permissions
deny (0)
—
ask (0)
—
allow (2)
Bash(avdmanager list *)
Bash(brew list*)
Similar rigs
smg99/ticket-flow
A repeatable, high-discipline workflow pattern for Cursor agents: fetches issues, branches, implements, adds tests, runs human confirmation gates, commits, PRs, and notifies chat.
Minimalist 1.1k tok · · Q5
ibrahimkobeissy/harneo
Harness manager for Claude Code: one audited library of skills, subagents, settings and MCP servers, shipped into any project as a profile. Every write previewed and backed up.
Fort Knox 1.7k tok · · Q77
nagisanzenin/engram
Evidence-based learning engine — first-principles curricula, free-recall verification with receipts, FSRS-scheduled memory, and explorable artifacts. Learn anything; keep it.
Minimalist 185 tok · · Q0
jovesun-lab/arcgram
Arcgram — an interactive diagram engine for human–AI collaboration: an AI agent draws its reasoning as an auditable node-and-edge flow and self-audits it, delivered as one standalone HTML file. (Arcgram the AI diagram tool — not Archigram.)
Minimalist 31 tok · · Q0