tuyenht/Bearingkit
Open-source workflow for AI coding agents: routes each request to the right step, stops for approval before risky changes, and shows evidence before claiming done. Claude Code and Antigravity today; Gemini CLI, Cursor, Codex, Copilot CLI an
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 tuyenht/Bearingkit/agents ./rig-bearingkit/agents $ npx degit tuyenht/Bearingkit/skills ./rig-bearingkit/skills $ npx degit tuyenht/Bearingkit/hooks ./rig-bearingkit/hooks
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 tuyenht/Bearingkit/skills/bk-audit .claude/skills/bk-audit $ npx degit tuyenht/Bearingkit/skills/bk-build .claude/skills/bk-build $ npx degit tuyenht/Bearingkit/skills/bk-close .claude/skills/bk-close $ npx degit tuyenht/Bearingkit/skills/bk-db .claude/skills/bk-db $ npx degit tuyenht/Bearingkit/skills/bk-debug .claude/skills/bk-debug $ npx degit tuyenht/Bearingkit/skills/bk-design .claude/skills/bk-design $ npx degit tuyenht/Bearingkit/skills/bk-map .claude/skills/bk-map $ npx degit tuyenht/Bearingkit/skills/bk-next .claude/skills/bk-next $ npx degit tuyenht/Bearingkit/skills/bk-ops .claude/skills/bk-ops $ npx degit tuyenht/Bearingkit/skills/bk-perf .claude/skills/bk-perf $ npx degit tuyenht/Bearingkit/skills/bk-plan .claude/skills/bk-plan $ npx degit tuyenht/Bearingkit/skills/bk-protocol .claude/skills/bk-protocol $ npx degit tuyenht/Bearingkit/skills/bk-research .claude/skills/bk-research $ npx degit tuyenht/Bearingkit/skills/bk-review .claude/skills/bk-review $ npx degit tuyenht/Bearingkit/skills/bk-setup .claude/skills/bk-setup $ npx degit tuyenht/Bearingkit/skills/bk-ship .claude/skills/bk-ship $ npx degit tuyenht/Bearingkit/skills/bk-spec .claude/skills/bk-spec $ npx degit tuyenht/Bearingkit/skills/bk-test .claude/skills/bk-test
$ curl -fsSL --create-dirs -o .claude/agents/bk-design-critic.md https://raw.githubusercontent.com/tuyenht/Bearingkit/main/agents/bk-design-critic.md $ curl -fsSL --create-dirs -o .claude/agents/bk-query-optimizer.md https://raw.githubusercontent.com/tuyenht/Bearingkit/main/agents/bk-query-optimizer.md $ curl -fsSL --create-dirs -o .claude/agents/bk-researcher.md https://raw.githubusercontent.com/tuyenht/Bearingkit/main/agents/bk-researcher.md $ curl -fsSL --create-dirs -o .claude/agents/bk-reviewer.md https://raw.githubusercontent.com/tuyenht/Bearingkit/main/agents/bk-reviewer.md
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 7,204 rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(rm -rf *)",
"Read(**/*.pem)",
"Bash(rm -rf /)",
"Bash(git push --force:*)",
"Bash(sudo *)",
"Read(.env)",
"Bash(rm -rf /*)",
"Read(./.env.*)",
"Read(~/.aws/**)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Read(**/*.key)",
"Read(**/.env.*)",
"Bash(sudo:*)",
"Bash(git reset --hard*)",
"Bash(git reset --hard:*)",
"Read(.env.*)"
],
"ask": [
"Bash(git push:*)",
"Bash(git push *)",
"Bash(git commit:*)",
"Bash(rm *)",
"Bash(rm:*)",
"Bash(git rebase *)",
"Bash(wget *)",
"Bash(npm publish:*)",
"Bash(git commit *)",
"Bash(gh pr merge *)"
]
}
} Skills (18)
bk-auditbk-buildbk-closebk-dbbk-debugbk-designbk-mapbk-nextbk-opsbk-perfbk-planbk-protocolbk-researchbk-reviewbk-setupbk-shipbk-specbk-test
Subagents (4)
| bk-design-critic model: sonnet | Reviews a UI change for generic defaults and accessibility gaps (contrast, focus states, reduced motion, tokens). For bk-design, bk-review. |
| bk-query-optimizer model: sonnet | Diagnoses a slow query from its EXPLAIN plan (logical reads, plan shape), read-only, never printing credentials. For bk-db and bk-perf. |
| bk-researcher model: sonnet | Answers a technical question a skill cannot verify locally, from independent sources and the pinned major's docs, with confidence labels. |
| bk-reviewer model: opus | Independent review of hot-path changes (auth, payments, data deletion, migrations, tenancy) by a non-author. For bk-review and bk-ship. |
Similar rigs
obra/superpowers
An agentic skills framework & software development methodology that works.
Skill Collector 3.0k tok ·
anthropics/skills
Public repository for Agent Skills
Skill Collector 1.3k tok ·
anthropics/claude-code
Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.
Skill Collector 1.4k tok ·
JuliusBrussee/caveman
🪨 why use many token when few token do trick. Viral skill + proxy for coding agents that cuts 65% of tokens by talking like a caveman.
Skill Collector 16.5k tok ·