abyss-node/claude-config
Global Claude Code configuration: rules, subagents and commands.
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 abyss-node/claude-config/agents ./rig-claude-config/agents $ npx degit abyss-node/claude-config/commands ./rig-claude-config/commands
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.
$ curl -fsSL --create-dirs -o .claude/agents/code-reviewer.md https://raw.githubusercontent.com/abyss-node/claude-config/master/agents/code-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/consultant.md https://raw.githubusercontent.com/abyss-node/claude-config/master/agents/consultant.md $ curl -fsSL --create-dirs -o .claude/agents/verifier.md https://raw.githubusercontent.com/abyss-node/claude-config/master/agents/verifier.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 *)"
]
}
} Subagents (3)
| code-reviewer model: opus | Reviews a diff against the ways this specific project breaks, not against general good practice. Use before any commit touching money, authorisation, time, migrations, or a contract another component |
| consultant model: sonnet | Answers a question about one part of a codebase on behalf of a session working in another part — what an endpoint returns, what consumes a field, whether a rename would break something, how a subsyste |
| verifier model: opus | Re-runs ground truth against a claim that work is finished — checks out the state, builds it, runs the suite, and reports what actually happened. Use on any work another agent reported as done, before |
Slash commands (2)
/checkpoint/lanes
Similar rigs
Remus3/Legion-Wallpaper
Self-auditing AI image restoration pipeline (super-resolution, LaMa inpainting, face repair, metric-plus-vision gate ladder), built by a multi-agent Claude Code system with independent verifiers, git-hook gates and headless background lanes
Automator 6.5k tok · · Q49
viberific/Autora
An agentic harness you can watch: the agent's browser, terminal, desktop and file edits stream into the conversation live, and every session records itself. Runs on Umbrel. Built with Claude Code.
Pragmatist 3.1k tok · · Q29
AlejoPH03/roundtrip-travel-app
Social travel app prototype: friends' trips on a 3D globe, trusted recommendations and group trip planning. Vanilla JS, no build step.
Minimalist 1.6k tok · · Q3
berni23/laight
Run teams of coding agents in tmux from YAML workflows, with one queue for every human decision.
Pragmatist 8.3k tok · · Q35