JDevlieghere/dotfiles
⚙ Dotfiles for Linux & macOS
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
Copy this rig
# review before running: this installs third-party code
$ npx degit JDevlieghere/dotfiles/.claude ./rig-dotfiles # 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.
$ npx degit JDevlieghere/dotfiles/.claude/skills/comments .claude/skills/comments $ npx degit JDevlieghere/dotfiles/.claude/skills/commit-message .claude/skills/commit-message $ npx degit JDevlieghere/dotfiles/.claude/skills/jsc-development .claude/skills/jsc-development $ npx degit JDevlieghere/dotfiles/.claude/skills/llvm-development .claude/skills/llvm-development $ npx degit JDevlieghere/dotfiles/.claude/skills/over-engineering .claude/skills/over-engineering $ npx degit JDevlieghere/dotfiles/.claude/skills/review-queue .claude/skills/review-queue
$ curl -fsSL --create-dirs -o .claude/agents/review-adversarial.md https://raw.githubusercontent.com/JDevlieghere/dotfiles/main/.claude/agents/review-adversarial.md $ curl -fsSL --create-dirs -o .claude/agents/review-api.md https://raw.githubusercontent.com/JDevlieghere/dotfiles/main/.claude/agents/review-api.md $ curl -fsSL --create-dirs -o .claude/agents/review-concurrency.md https://raw.githubusercontent.com/JDevlieghere/dotfiles/main/.claude/agents/review-concurrency.md $ curl -fsSL --create-dirs -o .claude/agents/review-conventions.md https://raw.githubusercontent.com/JDevlieghere/dotfiles/main/.claude/agents/review-conventions.md $ curl -fsSL --create-dirs -o .claude/agents/review-correctness.md https://raw.githubusercontent.com/JDevlieghere/dotfiles/main/.claude/agents/review-correctness.md $ curl -fsSL --create-dirs -o .claude/agents/review-design.md https://raw.githubusercontent.com/JDevlieghere/dotfiles/main/.claude/agents/review-design.md $ curl -fsSL --create-dirs -o .claude/agents/review-docs.md https://raw.githubusercontent.com/JDevlieghere/dotfiles/main/.claude/agents/review-docs.md $ curl -fsSL --create-dirs -o .claude/agents/review-performance.md https://raw.githubusercontent.com/JDevlieghere/dotfiles/main/.claude/agents/review-performance.md $ curl -fsSL --create-dirs -o .claude/agents/review-security.md https://raw.githubusercontent.com/JDevlieghere/dotfiles/main/.claude/agents/review-security.md $ curl -fsSL --create-dirs -o .claude/agents/review-tests.md https://raw.githubusercontent.com/JDevlieghere/dotfiles/main/.claude/agents/review-tests.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 (6)
Subagents (10)
| review-adversarial | Stress-tests a change against its own description — verifies every claim the author makes ("fixes X", "NFC", "safe because Y") against the code, probes the edges they didn't mention, and checks the di |
| review-api | Reviews a diff for public surface and compatibility — signature and behavior breaks, ABI hazards, serialized format and protocol changes, internals leaking through public headers. Use when a change to |
| review-concurrency | Reviews a diff for concurrency defects — data races, missing or excessive locking, lock-ordering deadlocks, atomicity gaps, unsafe lazy initialization, assumptions about which thread a callback runs o |
| review-conventions | Checks a diff against the project's own conventions by running its formatter and linter first, then hand-flagging only what the tools cannot see. Use when reviewing a PR or branch for naming, formatti |
| review-correctness | Reviews a diff for bugs — wrong logic, bad edge cases, unhandled errors, lifetime and null hazards, ordering and state mistakes. Use when reviewing a PR, a branch, or the working tree for defects. |
| review-design | Reviews a diff for design — architecture, layering, abstraction level, coupling, complexity added or left behind, type and container choices. Use when reviewing a PR or branch for shape rather than li |
| review-docs | Reviews a diff for documentation and comment quality — stale docs for changed behavior, missing docs on new public surface, and comments that restate the code, pin themselves to one example, or ramble |
| review-performance | Reviews a diff for performance problems — avoidable allocations and copies, the wrong container or algorithm for the access pattern, redundant work in a loop, repeated I/O. Use when a change touches a |
| review-security | Reviews a diff for security defects — untrusted input handling, injection, missing authorization, memory safety, secret exposure, unsafe file and path handling. Use when a change touches parsing, exte |
| review-tests | Reviews a diff for test coverage and test quality — whether new behavior is exercised, whether a fix has a regression test that fails without it, and whether the tests assert anything real. Use when r |
Similar rigs
trailofbits/coop
Isolated VM environment for running Claude Code and Codex
Orchestrator 2.7k tok ·
AlexisBalayre/agentspine
One shared agent setup for Claude Code, Codex, opencode, Mistral Vibe and Cursor: AGENTS.md, skills and blocking hooks stored once in .agents/ and wired into every tool.
Orchestrator 3.3k tok ·
AlexisBalayre/claude-code-config
Opinionated Claude Code config: path-scoped rules, 34 skills, 12 subagents, deterministic hooks, and a multi-agent PR review. Stack-agnostic; run /adapt-to-project to fit any repo.
Fort Knox 6.4k tok ·
netopsengineer/axiom
Axiom is a curated marketplace of shared plugins for Claude Code and Codex.
Pragmatist 3.0k tok ·