ku5ic/claude-kit
Ship with confidence. Hooks that make Claude Code back its claims and pass your project's own checks before it says "done", stack skills that load on their own, and 45 commands for the work around the code: audits, ADRs and RFCs, reviews of
field: unknown workflow: Security reviewFrontend & designPR / code reviewPlanning & specsTesting & TDD runs claude in CI
stack: Go
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
Copy this rig
# review before running: this installs third-party code
$ npx degit ku5ic/claude-kit/.claude ./rig-claude-kit # 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 ku5ic/claude-kit/.claude/skills/add-language .claude/skills/add-language $ npx degit ku5ic/claude-kit/skills/audit-a11y .claude/skills/audit-a11y $ npx degit ku5ic/claude-kit/skills/audit-arch .claude/skills/audit-arch $ npx degit ku5ic/claude-kit/skills/audit-ci .claude/skills/audit-ci $ npx degit ku5ic/claude-kit/skills/audit-debt .claude/skills/audit-debt $ npx degit ku5ic/claude-kit/skills/audit-doc-drift .claude/skills/audit-doc-drift $ npx degit ku5ic/claude-kit/skills/audit-flags .claude/skills/audit-flags $ npx degit ku5ic/claude-kit/skills/audit-infra .claude/skills/audit-infra $ npx degit ku5ic/claude-kit/skills/audit-perf .claude/skills/audit-perf $ npx degit ku5ic/claude-kit/skills/audit-readiness .claude/skills/audit-readiness $ npx degit ku5ic/claude-kit/skills/audit-security .claude/skills/audit-security $ npx degit ku5ic/claude-kit/skills/audit-slop .claude/skills/audit-slop $ npx degit ku5ic/claude-kit/skills/audit-tokens .claude/skills/audit-tokens $ npx degit ku5ic/claude-kit/skills/audit-verify .claude/skills/audit-verify $ npx degit ku5ic/claude-kit/skills/backup-patterns .claude/skills/backup-patterns $ npx degit ku5ic/claude-kit/skills/bash-patterns .claude/skills/bash-patterns $ npx degit ku5ic/claude-kit/skills/code-challenge .claude/skills/code-challenge $ npx degit ku5ic/claude-kit/skills/code-contract .claude/skills/code-contract $ npx degit ku5ic/claude-kit/skills/code-deep-review .claude/skills/code-deep-review $ npx degit ku5ic/claude-kit/skills/code-evaluate .claude/skills/code-evaluate $ npx degit ku5ic/claude-kit/skills/code-from-design .claude/skills/code-from-design $ npx degit ku5ic/claude-kit/skills/code-history .claude/skills/code-history $ npx degit ku5ic/claude-kit/skills/code-host-patterns .claude/skills/code-host-patterns $ npx degit ku5ic/claude-kit/skills/code-impact .claude/skills/code-impact $ npx degit ku5ic/claude-kit/skills/code-migrate .claude/skills/code-migrate $ npx degit ku5ic/claude-kit/skills/code-tour .claude/skills/code-tour $ npx degit ku5ic/claude-kit/skills/code-visual .claude/skills/code-visual $ npx degit ku5ic/claude-kit/skills/deps .claude/skills/deps $ npx degit ku5ic/claude-kit/skills/django-patterns .claude/skills/django-patterns $ npx degit ku5ic/claude-kit/skills/docker-patterns .claude/skills/docker-patterns $ npx degit ku5ic/claude-kit/skills/drf-patterns .claude/skills/drf-patterns $ npx degit ku5ic/claude-kit/skills/engineering-fundamentals .claude/skills/engineering-fundamentals $ npx degit ku5ic/claude-kit/skills/fastapi-patterns .claude/skills/fastapi-patterns $ npx degit ku5ic/claude-kit/skills/git-patterns .claude/skills/git-patterns $ npx degit ku5ic/claude-kit/skills/go-patterns .claude/skills/go-patterns $ npx degit ku5ic/claude-kit/skills/graphql-patterns .claude/skills/graphql-patterns $ npx degit ku5ic/claude-kit/skills/investigate .claude/skills/investigate $ npx degit ku5ic/claude-kit/skills/javascript-patterns .claude/skills/javascript-patterns $ npx degit ku5ic/claude-kit/skills/kit-rules .claude/skills/kit-rules $ npx degit ku5ic/claude-kit/skills/logging-patterns .claude/skills/logging-patterns
$ curl -fsSL --create-dirs -o .claude/agents/auditor.md https://raw.githubusercontent.com/ku5ic/claude-kit/main/agents/auditor.md $ curl -fsSL --create-dirs -o .claude/agents/claim-checker.md https://raw.githubusercontent.com/ku5ic/claude-kit/main/agents/claim-checker.md $ curl -fsSL --create-dirs -o .claude/agents/debugger.md https://raw.githubusercontent.com/ku5ic/claude-kit/main/agents/debugger.md $ curl -fsSL --create-dirs -o .claude/agents/gatherer.md https://raw.githubusercontent.com/ku5ic/claude-kit/main/agents/gatherer.md $ curl -fsSL --create-dirs -o .claude/agents/plan-critic.md https://raw.githubusercontent.com/ku5ic/claude-kit/main/agents/plan-critic.md $ curl -fsSL --create-dirs -o .claude/agents/researcher.md https://raw.githubusercontent.com/ku5ic/claude-kit/main/agents/researcher.md $ curl -fsSL --create-dirs -o .claude/agents/tester.md https://raw.githubusercontent.com/ku5ic/claude-kit/main/agents/tester.md
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 (77)
add-languageaudit-a11yaudit-archaudit-ciaudit-debtaudit-doc-driftaudit-flagsaudit-infraaudit-perfaudit-readinessaudit-securityaudit-slopaudit-tokensaudit-verifybackup-patternsbash-patternscode-challengecode-contractcode-deep-reviewcode-evaluatecode-from-designcode-historycode-host-patternscode-impactcode-migratecode-tourcode-visualdepsdjango-patternsdocker-patternsdrf-patternsengineering-fundamentalsfastapi-patternsgit-patternsgo-patternsgraphql-patternsinvestigatejavascript-patternskit-ruleslogging-patternsmeta-conventionsmeta-promptmeta-refreshmeta-retrometa-skillmonitoring-patternsnext-app-router-patternsnuxt-patternsopentofu-patternspr-descriptionpr-publishpr-queuepr-replypr-reviewpython-patternsreact-patternsreport-formatsecurity-patternsslop-patternsstorybook-patternstailwind-patternstest-patternstypescript-patternsunslopvps-provisioningvue-patternswcag-auditwrite-adrwrite-commitwrite-devnotewrite-explainerwrite-release-noteswrite-retrowrite-rfcwrite-runbookwrite-stakeholderwrite-tickets
Subagents (7)
| auditor | Read-only audit of a code surface for accessibility, performance, architecture, technical debt, documentation drift, security, CI, infrastructure, or prose slop. The invoking audit-* skill supplies th |
| claim-checker model: sonnet | Checks the factual claims in a text before anyone relies on it, the way a reviewer who trusts nothing would. A hook asks for it before an investigation, review, `audit-*`, or `code-*` report is writte |
| debugger | Investigates unexpected behavior to localize a fault, using read access plus at most one targeted probe edit. Never applies the fix. Use to find where and why something breaks; hand the fix back to th |
| gatherer model: haiku | Read-only fact gathering for one question from the investigate skill's gatherer list (precedent, tests, history). Returns facts with file:line, never opinions or advice. Use when skills/investigate na |
| plan-critic model: opus | Adversarially reviews a plan file (plan mode output) against the real repo, not only the plan's own internal consistency. Checks that cited precedent says what the plan claims, catches design-integrit |
| researcher model: sonnet | Read-only external research in primary sources (official docs, changelogs, release notes, package source, language specs) via WebSearch and WebFetch. Delegate version checks, API verification, and doc |
| tester | Adds or updates tests for recent implementation work, then runs them. Never alters implementation code to make a test pass. Use when a change needs test coverage or existing tests need extending. |
Similar rigs
Izangi2714/claude-code-python-stack
Boost Claude Code for Python with ready-made rules, tests, and QA automation for consistent, secure development
Orchestrator 3.3k tok · · Q5
xaaha/dev-env
Neovim Config and other dot-files.
YOLO Cowboy 1.7k tok · · Q5
LucasXCoder/noetra-claude-setup
Tuned Claude Code setup: mandatory verify-before-done workflow, 84 skills, 33 agents, guard hooks, status line, cost-aware settings, and a one-command Mac installer.
Skill Collector 7.4k tok · · Q5
komluk/scaffolding
Spec-driven multi-agent orchestration plugin for Claude Code — 13 agents, 35 skills, per-phase model tiers, opt-in lifecycle hooks, and optional cross-device semantic memory.
YOLO Cowboy 5.2k tok · · Q11