code-yeongyu/my-claude-code-harness
I'm confident this setup will become deprecated and outdated in less than 3 months
ARCHETYPE
Pragmatist
A balanced, no-drama setup: some rules, some tools, nothing extreme.
Copy this rig
# review before running: this installs third-party code $ npx degit code-yeongyu/my-claude-code-harness/agents ./rig-my-claude-code-harness/agents $ npx degit code-yeongyu/my-claude-code-harness/commands ./rig-my-claude-code-harness/commands $ npx degit code-yeongyu/my-claude-code-harness/skills ./rig-my-claude-code-harness/skills $ npx degit code-yeongyu/my-claude-code-harness/hooks ./rig-my-claude-code-harness/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 code-yeongyu/my-claude-code-harness/skills/git-committer .claude/skills/git-committer $ npx degit code-yeongyu/my-claude-code-harness/skills/pr-context-retrieval .claude/skills/pr-context-retrieval $ npx degit code-yeongyu/my-claude-code-harness/skills/pr-creator .claude/skills/pr-creator
$ curl -fsSL --create-dirs -o .claude/agents/executor.md https://raw.githubusercontent.com/code-yeongyu/my-claude-code-harness/master/agents/executor.md $ curl -fsSL --create-dirs -o .claude/agents/librarian.md https://raw.githubusercontent.com/code-yeongyu/my-claude-code-harness/master/agents/librarian.md $ curl -fsSL --create-dirs -o .claude/agents/plan-reviewer.md https://raw.githubusercontent.com/code-yeongyu/my-claude-code-harness/master/agents/plan-reviewer.md
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 6,974 rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(rm -rf *)",
"Read(**/*.pem)",
"Bash(rm -rf /)",
"Bash(git push --force:*)",
"Bash(sudo *)",
"Read(~/.aws/**)",
"Bash(rm -rf /*)",
"Read(./.env.*)",
"Read(.env)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Read(**/.env.*)",
"Read(**/*.key)",
"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(npm publish:*)",
"Bash(wget *)",
"Bash(git rebase *)",
"Bash(gh pr merge *)",
"Bash(git commit *)"
]
}
} Skills (3)
Subagents (3)
| executor model: haiku | MUST BE USED when executing tasks from ai-todo list plans. Executes EXACTLY ONE TODO checkbox item and meticulously records insights in Notepad section. |
| librarian model: haiku | Specialized research agent for searching remote codebases, retrieving official documentation, and finding implementation examples using GitHub CLI and Context7. MUST BE USED when users ask to look up |
| plan-reviewer | Expert who reviews whether work plans contain sufficient information according to the set detail level. Evaluates using different criteria for each level. |
Hooks (3)
| event | matcher | runs |
|---|---|---|
| UserPromptSubmit | * | ~/.claude/hooks/user-prompt-submit/system-reminder.py |
| PostToolUse | Bash|Write|Edit|MultiEdit|Read|Task | uv run ~/.claude/hooks/post_tool_use.py |
| Stop | * | uv run --script ~/.claude/hooks/stop.py |
Slash commands (5)
/commit/create-pr/execute/planner/read-pr
Similar rigs
HGInsights/claude-code-setup
Production-tested Claude Code harness configuration: hooks, skills, subagents, and automated review loops
Pragmatist 1.9k tok ·
Dana-Afazeli/agentic-swe-kit
A maintained template for the agentic-SWE harness: proven gates, Claude Code hooks, CI, a headless review loop, and the workflow docs behind them. Use this template, run kit.py init, take updates with kit.py update.
Fort Knox 1.7k tok ·
andrewhampton/dotfiles
My dotfiles
Minimalist 1.3k tok ·
RemiAsselin42/claude-config
Memory, features, skills, and agents for Claude, centralized and versioned, one-command installation
Automator 2.1k tok ·