~ / rigs / pkulijing / coding-agent-config

pkulijing/coding-agent-config

My personal claude code coding workflow

↗ GitHub ★ 0 no license updated 11d ago personal setup Claude CodeCodex Claude plugin
field: unknown workflow: Git & commitsDocsPR / code review
stack: Python
share on X
ARCHETYPE
Skill Collector
Ten-plus skills loaded on demand. A procedural-knowledge library.
CONTEXT TAX · EVERY TURN
~1.2k tokens
Featherweight · median rig: 2.3k · breakdown
GUARDRAILS
0/5
No committed guardrails · details
Quality 0/100 Guardrails v2

Copy this rig

# review before running: this installs third-party code
$ npx degit pkulijing/coding-agent-config/agents ./rig-coding-agent-config/agents
$ npx degit pkulijing/coding-agent-config/skills ./rig-coding-agent-config/skills
$ npx degit pkulijing/coding-agent-config/hooks ./rig-coding-agent-config/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 pkulijing/coding-agent-config/skills/backlog .claude/skills/backlog
$ npx degit pkulijing/coding-agent-config/skills/bootstrap .claude/skills/bootstrap
$ npx degit pkulijing/coding-agent-config/skills/commit .claude/skills/commit
$ npx degit pkulijing/coding-agent-config/skills/design .claude/skills/design
$ npx degit pkulijing/coding-agent-config/skills/devtree .claude/skills/devtree
$ npx degit pkulijing/coding-agent-config/skills/finish .claude/skills/finish
$ npx degit pkulijing/coding-agent-config/skills/paper-read .claude/skills/paper-read
$ npx degit pkulijing/coding-agent-config/skills/pybump .claude/skills/pybump
$ npx degit pkulijing/coding-agent-config/skills/quick .claude/skills/quick
$ npx degit pkulijing/coding-agent-config/skills/rebase .claude/skills/rebase
$ npx degit pkulijing/coding-agent-config/skills/review-loop .claude/skills/review-loop
$ npx degit pkulijing/coding-agent-config/skills/routine-dev .claude/skills/routine-dev
$ npx degit pkulijing/coding-agent-config/skills/routine-slim .claude/skills/routine-slim
$ npx degit pkulijing/coding-agent-config/skills/start .claude/skills/start
$ npx degit pkulijing/coding-agent-config/skills/sync-project-config .claude/skills/sync-project-config
$ npx degit pkulijing/coding-agent-config/skills/triage .claude/skills/triage
$ npx degit pkulijing/coding-agent-config/templates/cli-plugin/__root__/plugin/skills/example-status .claude/skills/example-status
$ curl -fsSL --create-dirs -o .claude/agents/code-reviewer-deep.md https://raw.githubusercontent.com/pkulijing/coding-agent-config/master/agents/code-reviewer-deep.md
$ curl -fsSL --create-dirs -o .claude/agents/code-reviewer.md https://raw.githubusercontent.com/pkulijing/coding-agent-config/master/agents/code-reviewer.md
$ curl -fsSL --create-dirs -o .claude/agents/review-orchestrator.md https://raw.githubusercontent.com/pkulijing/coding-agent-config/master/agents/review-orchestrator.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 (17)

Subagents (3)

code-reviewer-deep
model: opus
并发 / 状态机 / 资源生命周期专项深审 reviewer,跑 opus,仅重档使用。只读不写、不再扇出子 agent。由 /review-loop 的 orchestrator 起。
code-reviewer
model: sonnet
单角度 code reviewer。按分配到的角度清单审一份 diff,返回带 file:line 证据的 finding 列表。只读不写、不再扇出子 agent。由 /review-loop 的 orchestrator 起。
review-orchestrator
model: sonnet
/review-loop 的编队 orchestrator。并行起 reviewer 子 agent 各审一个角度,跨 reviewer 去重、置信打分、探针验证,返回单一 finding 列表。只读不写。

Similar rigs

copied ✓