rochaksharma19/claude-code-setup
Production-grade Claude Code configuration stack for AI/ML development
ARCHETYPE
Pragmatist
A balanced, no-drama setup: some rules, some tools, nothing extreme.
Copy this rig
# review before running: this installs third-party code $ claude mcp add vexp -e VEXP_MEMORY_DIR=YOUR_VEXP_MEMORY_DIR -e VEXP_PROJECT=YOUR_VEXP_PROJECT -- npx -y @vexp/mcp-server@latest $ claude mcp add github -e GITHUB_PERSONAL_ACCESS_<redacted> -- npx -y @modelcontextprotocol/server-github $ claude mcp add filesystem -- npx -y @modelcontextprotocol/server-filesystem ${HOME}/your-projects-path $ claude mcp add brave-search -e BRAVE_<redacted> -- npx -y @modelcontextprotocol/server-brave-search $ claude mcp add context7 -- npx -y @upstash/context7-mcp@latest $ npx degit rochaksharma19/claude-code-setup/.claude ./rig-claude-code-setup # 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.
$ claude mcp add vexp -e VEXP_MEMORY_DIR=YOUR_VEXP_MEMORY_DIR -e VEXP_PROJECT=YOUR_VEXP_PROJECT -- npx -y @vexp/mcp-server@latest $ claude mcp add github -e GITHUB_PERSONAL_ACCESS_<redacted> -- npx -y @modelcontextprotocol/server-github $ claude mcp add filesystem -- npx -y @modelcontextprotocol/server-filesystem ${HOME}/your-projects-path $ claude mcp add brave-search -e BRAVE_<redacted> -- npx -y @modelcontextprotocol/server-brave-search $ claude mcp add context7 -- npx -y @upstash/context7-mcp@latest
$ npx degit rochaksharma19/claude-code-setup/.claude/skills/claude-pr-checklist .claude/skills/claude-pr-checklist $ npx degit rochaksharma19/claude-code-setup/.claude/skills/new-rag-eval .claude/skills/new-rag-eval
$ curl -fsSL --create-dirs -o .claude/agents/eval-runner.md https://raw.githubusercontent.com/rochaksharma19/claude-code-setup/main/.claude/agents/eval-runner.md $ curl -fsSL --create-dirs -o .claude/agents/prompt-auditor.md https://raw.githubusercontent.com/rochaksharma19/claude-code-setup/main/.claude/agents/prompt-auditor.md $ curl -fsSL --create-dirs -o .claude/agents/retrieval-reviewer.md https://raw.githubusercontent.com/rochaksharma19/claude-code-setup/main/.claude/agents/retrieval-reviewer.md
Merge into .claude/settings.json (project) or ~/.claude/settings.json (user). Hook commands reference scripts in the source repo — copy those too.
{
"hooks": {
"PreToolUse": [
{
"matcher": "Bash",
"hooks": [
{
"type": "command",
"command": "~/.claude/hooks/gate_git_push.sh"
}
]
}
]
}
} MCP servers (5)
| server | source | est. tokens |
|---|---|---|
| @vexp/mcp-server · "vexp" env: VEXP_MEMORY_DIR, VEXP_PROJECT | npm | 2.5k |
| GitHub MCP · "github" env: GITHUB_PERSONAL_ACCESS_TOKEN | npm | 18.0k |
| Filesystem | npm | 4.2k |
| Brave Search · "brave-search" env: BRAVE_API_KEY | npm | 900 |
| Context7 | npm | 1.2k |
Skills (2)
Subagents (3)
| eval-runner model: sonnet | Designs and implements evaluation frameworks for LLM applications. Use when setting up evals, defining metrics, or analyzing model output quality for agents, RAG pipelines, or classifiers. |
| prompt-auditor model: sonnet | Audits LLM prompts for clarity, robustness, and common failure modes. Use when writing or reviewing system prompts, few-shot examples, or tool descriptions for any LLM application. |
| retrieval-reviewer model: sonnet | Reviews RAG and retrieval pipeline changes for chunking correctness, reranker usage, and citation-contract regressions. Read-only. Invoke before opening a PR that touches retrieval code. |
Hooks (3)
| event | matcher | runs |
|---|---|---|
| PreToolUse | Bash | ~/.claude/hooks/gate_git_push.sh |
| PostToolUse | Write|Edit|MultiEdit | [ -n "$CLAUDE_TOOL_FILE_PATH" ] && [[ "$CLAUDE_TOOL_FILE_PATH" == *.py ]] && uv run ruff format "$CLAUDE_TOOL_FILE_PATH" >/dev/null 2>&1 || true |
| PermissionDenied | * | mkdir -p ~/.claude/logs && jq -c . >> ~/.claude/logs/denied.jsonl |
Similar rigs
cassler/awesome-claude-code-setup
Give Claude the super powers it deserves with this power user collection of bash scripts and slash commands
Pragmatist 31.8k tok ·
msbaek/dotfiles
—
Pragmatist 37.3k tok ·
floating-astronaut/vibe-coding-kit
Replace your dev team: an operating system for a coordinated Claude Code, Codex and Kimi AI coding team. Docs-first lanes, zero-drift handoffs. MIT.
Pragmatist 22.8k tok ·
vorburger/dotfiles
https://dotfiles.github.io
Pragmatist 22.0k tok ·