shimo4228/akc-cycle
Keeps the rules and skills in a Claude Code setup from piling up and going stale: a Claude Code plugin of audit and upkeep skills plus an optional always-loaded rules file, the installable form of the Agent Knowledge Cycle (AKC).
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 shimo4228/akc-cycle/skills ./rig-akc-cycle/skills 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 shimo4228/akc-cycle/skills/agent-stocktake .claude/skills/agent-stocktake $ npx degit shimo4228/akc-cycle/skills/author-calibrated-eval .claude/skills/author-calibrated-eval $ npx degit shimo4228/akc-cycle/skills/context-sync .claude/skills/context-sync $ npx degit shimo4228/akc-cycle/skills/decision-history .claude/skills/decision-history $ npx degit shimo4228/akc-cycle/skills/generation-audit .claude/skills/generation-audit $ npx degit shimo4228/akc-cycle/skills/harness-boundary .claude/skills/harness-boundary $ npx degit shimo4228/akc-cycle/skills/harness-stocktake .claude/skills/harness-stocktake $ npx degit shimo4228/akc-cycle/skills/hook-stocktake .claude/skills/hook-stocktake $ npx degit shimo4228/akc-cycle/skills/jev-judgment-design .claude/skills/jev-judgment-design $ npx degit shimo4228/akc-cycle/skills/learn-eval .claude/skills/learn-eval $ npx degit shimo4228/akc-cycle/skills/llm-as-judge .claude/skills/llm-as-judge $ npx degit shimo4228/akc-cycle/skills/repo-asset-stocktake .claude/skills/repo-asset-stocktake $ npx degit shimo4228/akc-cycle/skills/review-to-lint .claude/skills/review-to-lint $ npx degit shimo4228/akc-cycle/skills/rules-distill .claude/skills/rules-distill $ npx degit shimo4228/akc-cycle/skills/rules-stocktake .claude/skills/rules-stocktake $ npx degit shimo4228/akc-cycle/skills/search-first .claude/skills/search-first $ npx degit shimo4228/akc-cycle/skills/skill-comply .claude/skills/skill-comply $ npx degit shimo4228/akc-cycle/skills/skill-health .claude/skills/skill-health $ npx degit shimo4228/akc-cycle/skills/skill-stocktake .claude/skills/skill-stocktake $ npx degit shimo4228/akc-cycle/skills/skill-writer .claude/skills/skill-writer $ npx degit shimo4228/akc-cycle/skills/verify-bootstrap .claude/skills/verify-bootstrap
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 (21)
agent-stocktakeauthor-calibrated-evalcontext-syncdecision-historygeneration-auditharness-boundaryharness-stocktakehook-stocktakejev-judgment-designlearn-evalllm-as-judgerepo-asset-stocktakereview-to-lintrules-distillrules-stocktakesearch-firstskill-complyskill-healthskill-stocktakeskill-writerverify-bootstrap
Similar rigs
shimo4228/harness-stocktake
Audits your whole Claude Code harness (skills, rules, agents, hooks, CLAUDE.md, ADRs) for stale, contradictory or duplicated content, and changes only what you approve.
Skill Collector 721 tok · · Q3
lcleris/dotfiles
Loan CLERIS | dotfiles config
Automator 87 tok · · Q15
grvermeulen/agent-workflow-template
Personal template for AI-assisted development: AGENTS.md preferences, CodeRabbit config, Cursor rules+skills+hooks, Claude Code loop-on-ci command, husky hooks, and agent-flow GitHub workflows.
Pragmatist 1.7k tok · · Q3
rusharlabs/house-party-protocol
Evidence-first harness for coding agents across Claude Code and Codex CLI: cross-model review, governed loops, isolated lanes, deterministic evaluation, verified delivery.
Skill Collector 2.6k tok · · Q5