ksdisch/claude-config
Dotfiles for Claude Code — my version-controlled slash commands, skills, subagents, and global instructions, symlinked into ~/.claude
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 ksdisch/claude-config/.claude ./rig-claude-config # 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 ksdisch/claude-config/skills/adversarial-review .claude/skills/adversarial-review $ npx degit ksdisch/claude-config/skills/architecture-viewer .claude/skills/architecture-viewer $ npx degit ksdisch/claude-config/skills/audio-series .claude/skills/audio-series $ npx degit ksdisch/claude-config/skills/bug-hunt .claude/skills/bug-hunt $ npx degit ksdisch/claude-config/skills/cc-yt-idea-mine .claude/skills/cc-yt-idea-mine $ npx degit ksdisch/claude-config/skills/day .claude/skills/day $ npx degit ksdisch/claude-config/skills/family-compare .claude/skills/family-compare $ npx degit ksdisch/claude-config/skills/gauntlet .claude/skills/gauntlet $ npx degit ksdisch/claude-config/skills/kickoff .claude/skills/kickoff $ npx degit ksdisch/claude-config/skills/mock-call .claude/skills/mock-call $ npx degit ksdisch/claude-config/skills/narrate .claude/skills/narrate $ npx degit ksdisch/claude-config/skills/nlm-skill .claude/skills/nlm-skill $ npx degit ksdisch/claude-config/skills/notebook-init .claude/skills/notebook-init $ npx degit ksdisch/claude-config/skills/notebook-merge .claude/skills/notebook-merge $ npx degit ksdisch/claude-config/skills/onboard .claude/skills/onboard $ npx degit ksdisch/claude-config/skills/orchestrate .claude/skills/orchestrate $ npx degit ksdisch/claude-config/skills/paper-eli5 .claude/skills/paper-eli5 $ npx degit ksdisch/claude-config/skills/paper-figures .claude/skills/paper-figures $ npx degit ksdisch/claude-config/skills/paper-gloss .claude/skills/paper-gloss $ npx degit ksdisch/claude-config/skills/project-guide .claude/skills/project-guide $ npx degit ksdisch/claude-config/skills/project-wiki .claude/skills/project-wiki $ npx degit ksdisch/claude-config/skills/reorient .claude/skills/reorient $ npx degit ksdisch/claude-config/skills/replenish .claude/skills/replenish $ npx degit ksdisch/claude-config/skills/research-paper .claude/skills/research-paper $ npx degit ksdisch/claude-config/skills/retire .claude/skills/retire $ npx degit ksdisch/claude-config/skills/reweave .claude/skills/reweave $ npx degit ksdisch/claude-config/skills/seed-hunt .claude/skills/seed-hunt $ npx degit ksdisch/claude-config/skills/ship-and-route .claude/skills/ship-and-route $ npx degit ksdisch/claude-config/skills/teach-research .claude/skills/teach-research $ npx degit ksdisch/claude-config/skills/video-series .claude/skills/video-series $ npx degit ksdisch/claude-config/skills/visual-summary .claude/skills/visual-summary $ npx degit ksdisch/claude-config/skills/wayfinder-map .claude/skills/wayfinder-map $ npx degit ksdisch/claude-config/skills/youtube-breakdown .claude/skills/youtube-breakdown $ npx degit ksdisch/claude-config/skills/youtube-transcript .claude/skills/youtube-transcript
$ curl -fsSL --create-dirs -o .claude/agents/adversarial-reviewer.md https://raw.githubusercontent.com/ksdisch/claude-config/main/agents/adversarial-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/gauntlet-coder.md https://raw.githubusercontent.com/ksdisch/claude-config/main/agents/gauntlet-coder.md $ curl -fsSL --create-dirs -o .claude/agents/mutation-hardener.md https://raw.githubusercontent.com/ksdisch/claude-config/main/agents/mutation-hardener.md $ curl -fsSL --create-dirs -o .claude/agents/review-judge.md https://raw.githubusercontent.com/ksdisch/claude-config/main/agents/review-judge.md $ curl -fsSL --create-dirs -o .claude/agents/silent-failure-hunter.md https://raw.githubusercontent.com/ksdisch/claude-config/main/agents/silent-failure-hunter.md $ curl -fsSL --create-dirs -o .claude/agents/spec-miner.md https://raw.githubusercontent.com/ksdisch/claude-config/main/agents/spec-miner.md $ curl -fsSL --create-dirs -o .claude/agents/specifier.md https://raw.githubusercontent.com/ksdisch/claude-config/main/agents/specifier.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 (34)
adversarial-reviewarchitecture-vieweraudio-seriesbug-huntcc-yt-idea-minedayfamily-comparegauntletkickoffmock-callnarratenlm-skillnotebook-initnotebook-mergeonboardorchestratepaper-eli5paper-figurespaper-glossproject-guideproject-wikireorientreplenishresearch-paperretirereweaveseed-huntship-and-routeteach-researchvideo-seriesvisual-summarywayfinder-mapyoutube-breakdownyoutube-transcript
Subagents (7)
| adversarial-reviewer model: opus | Zero-context adversarial diff reviewer for the adversarial-review skill. Reads a branch's diff against its merge-base with fresh eyes and writes numbered, severity-graded findings (critical / should-f |
| gauntlet-coder model: opus | Builds one story to a written Gherkin specification — unit tests plus the implementation that satisfies them — running the repo's suite locally as it works and leaving the tree dirty for the orchestra |
| mutation-hardener model: sonnet | Dual-mode mutation-testing agent. `AUDIT` mode runs the mutation tool over a scope and reports surviving mutants graded on bug-hunt's critical/high/medium/low rubric, editing nothing. `HARDEN` mode ad |
| review-judge model: opus | Neutral zero-context judge for the adversarial-review skill. Rules ONLY on findings the author disputed — reads the reviewer's claim, the author's rebuttal, and the actual code at HEAD, then rules UPH |
| silent-failure-hunter model: sonnet | Read-only auditor that hunts silent failures — swallowed errors, empty catch blocks, dangerous fallbacks, broken error propagation, and missing boundary handling — over a given scope (path, module, or |
| spec-miner model: opus | Extracts behavioral specs from an existing codebase into flat Requirement/Invariant blocks with machine-parseable metadata (entities, enforced, id, test anchors), one spec file per capability at opens |
| specifier model: opus | Turns a story into an executable-shaped specification — a Gherkin acceptance test (`.feature`) plus a human-POV QA procedure — grounded in the target repo's existing behavior and conventions. Writes e |
Slash commands (17)
/begin/boot_server/brainstorm/catchup/claudify-repo/crap-check/envsetup/explore-plan/handoff-session/launch/prompt-optimize/smoke-test/tdd-loop/trim-context/wiki-backfill/wiki-init/wrap
Similar rigs
Chillin-spree/claude-code-system-template
A lean, skills-first Claude Code template for building web apps — vanilla by default, stack-swappable, with workflow skills, agents, and a commit/push safety gate.
Fort Knox 1.5k tok ·
VonTerraProject501c3/slushpile
An adversarial job search that gets better as it goes: pool-anchored fit scoring, a resume and cover letter built against the job description, seven agents that try to reject you before a recruiter does, and a profile that keeps what they f
Orchestrator 12.9k tok ·
poshan0126/dotclaude
The standard .claude/ folder structure for everyday development.
Fort Knox 1.7k tok ·
swingerman/engineer
Disciplined Agentic Engineering — a methodology kit for Claude Code: acceptance-test-first specs, explicit checkpoints, and autonomy you can actually leave running. The engineer, atdd and crap-analyzer plugins.
Skill Collector 1.8k tok ·