agentrigs
+ submit rig
~ / rigs / Kalabint / llm-tooling

Kalabint/llm-tooling

AI-Tooling Sammelsurium

↗ GitHub ★ 6 no license updated 1mo ago project Claude Code
share on X
ARCHETYPE
Skill Collector
Ten-plus skills loaded on demand. A procedural-knowledge library.
CONTEXT TAX · EVERY TURN
~1.0k tokens
Featherweight · median rig: 2.3k · breakdown
GUARDRAILS
0/5
No committed guardrails · details

Copy this rig

# review before running: this installs third-party code
$ npx degit Kalabint/llm-tooling/.claude ./rig-llm-tooling  # 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 Kalabint/llm-tooling/.claude/skills/ask-when-unclear .claude/skills/ask-when-unclear
$ npx degit Kalabint/llm-tooling/.claude/skills/calibrated-inference .claude/skills/calibrated-inference
$ npx degit Kalabint/llm-tooling/.claude/skills/check-dont-assume .claude/skills/check-dont-assume
$ npx degit Kalabint/llm-tooling/.claude/skills/code-is-authority .claude/skills/code-is-authority
$ npx degit Kalabint/llm-tooling/.claude/skills/discriminator .claude/skills/discriminator
$ npx degit Kalabint/llm-tooling/.claude/skills/handoff .claude/skills/handoff
$ npx degit Kalabint/llm-tooling/.claude/skills/memory-audit .claude/skills/memory-audit
$ npx degit Kalabint/llm-tooling/.claude/skills/no-flip-flop .claude/skills/no-flip-flop
$ npx degit Kalabint/llm-tooling/.claude/skills/no-invented-hazards .claude/skills/no-invented-hazards
$ npx degit Kalabint/llm-tooling/.claude/skills/no-self-report-omissions .claude/skills/no-self-report-omissions
$ npx degit Kalabint/llm-tooling/.claude/skills/no-sycophancy .claude/skills/no-sycophancy
$ npx degit Kalabint/llm-tooling/.claude/skills/no-unmeasured-numbers .claude/skills/no-unmeasured-numbers
$ npx degit Kalabint/llm-tooling/.claude/skills/own-guardrails-bind-me .claude/skills/own-guardrails-bind-me
$ npx degit Kalabint/llm-tooling/.claude/skills/prepare-next-steps .claude/skills/prepare-next-steps
$ npx degit Kalabint/llm-tooling/.claude/skills/smallest-clean-diff .claude/skills/smallest-clean-diff
$ npx degit Kalabint/llm-tooling/.claude/skills/solve-root-not-workaround .claude/skills/solve-root-not-workaround
$ npx degit Kalabint/llm-tooling/.claude/skills/surface-assumptions .claude/skills/surface-assumptions
$ npx degit Kalabint/llm-tooling/.claude/skills/untrusted-content-isolation .claude/skills/untrusted-content-isolation
$ npx degit Kalabint/llm-tooling/.claude/skills/verify-outcome .claude/skills/verify-outcome

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 (19)

ask-when-unclearcalibrated-inferencecheck-dont-assumecode-is-authoritydiscriminatorhandoffmemory-auditno-flip-flopno-invented-hazardsno-self-report-omissionsno-sycophancyno-unmeasured-numbersown-guardrails-bind-meprepare-next-stepssmallest-clean-diffsolve-root-not-workaroundsurface-assumptionsuntrusted-content-isolationverify-outcome

Similar rigs

marinasofia/sakura-terminal
A cherry blossom setup for Ghostty and Claude Code, with a flower pet that shows what Claude is doing
★ 3 project Claude Code 1 skills 6d ago
Minimalist 255 tok ·
epicurean-Paradox/claude-code-mastery
Battle-tested system for consistent, high-quality Claude Code sessions. Layered config, session continuity, decay prevention, skill-driven workflows.
★ 1 project Claude Code 1 skills 3d ago
Pragmatist 6.4k tok ·
VedantShirgaonkar/lodestone
Stop paying Claude Code's cache tax. Live usage and cache-expiry monitoring, automatic context handoffs, and one-command account switching. Cuts the token cost of switching accounts, cache expiry, limit resets, and bloated sessions. CLI plu
★ 0 project Claude Code 1 skills 4d ago
Minimalist 1.5k tok ·
jinmang2/ai-agent-dotfiles
Claude Code · Codex CLI · tmux · shell 설정. 훅/서브에이전트/플러그인을 직접 다루기 위한 dotfiles.
★ 0 setup Gemini CLI 1 skills 29d ago
Minimalist 97 tok ·
copied ✓