Manual-only skill for minimizing total codebase size. Only activate when explicitly requested by user. Measures success by final code amount, not effort. Bias toward deletion.
A manual-only skill for minimizing total codebase size by measuring success through final code amount, not effort. Biases toward deletion and aggressive simplification.
This skill exists to counter the natural tendency of codebases to grow over time. It provides a framework for evaluating changes based on whether they reduce the total amount of code in the final codebase, not just minimize the work required now.
More code creates more entropy. More entropy means more bugs, more maintenance, more cognitive load. This skill fights that trend.
Use this skill when:
Do NOT use automatically. This is a manual-only skill that requires explicit user request.
Measure the end state, not the effort.
Every change must answer:
What's the smallest codebase that solves this?
Does this result in less total code?
What can we delete?
Before starting, you must load at least one mindset from the references/ directory. These provide philosophical grounding for radical simplification:
references/Do not proceed without loading a mindset.
This skill actively rejects these common arguments:
| Argument | Why It's Rejected | |----------|-------------------| | "Keep what exists" | Status quo bias. The question is total code, not churn. | | "This adds flexibility" | Flexibility for what? YAGNI. | | "Better separation of concerns" | More files/functions = more code. Separation isn't free. | | "Type safety" | Worth how many lines? Sometimes runtime checks win. | | "Easier to understand" | 14 things are not easier than 2 things. |
Don't use this skill when:
User: "I need to add validation to these 5 forms"
Standard approach:
- Create validation schema for each form (5 files)
- Add validation helper functions (1 file)
- Update each form component (5 files modified)
- Total: +6 files, ~300 lines added
Reducing Entropy approach:
1. Load mindset (e.g., "worse-is-better.md")
2. Ask: What's the smallest codebase that solves this?
- Could we have 1 validation function instead of 5 schemas?
- Do all 5 forms need validation or just 2?
- Can we delete 3 forms and merge functionality?
3. Count before/after:
- Before: 5 forms + 0 validation = 500 lines
- Option A: 5 forms + 6 validation files = 800 lines (reject)
- Option B: 2 forms + 1 validation function = 350 lines (accept)
4. Delete: Remove 3 forms, add 1 function
- Net result: -150 lines
references/ - Philosophical mindsets to load before startingadding-reference-mindsets.md - How to add new mindsetsYou've succeeded when:
Bias toward deletion. Measure the end state.
Create or update AgentSkills. Use when designing, structuring, or packaging skills with scripts, references, and assets.
Delegate coding tasks to Codex, Claude Code, or Pi agents via background process. Use when: (1) building/creating new features or apps, (2) reviewing PRs (spawn in temp dir), (3) refactoring large codebases, (4) iterative coding that needs file exploration. NOT for: simple one-liner fixes (just edit), reading code (use read tool), thread-bound ACP harness requests in chat (for example spawn/run Codex or Claude Code in a Discord thread; use sessions_spawn with runtime:"acp"), or any work in ~/clawd workspace (never spawn agents here). Claude Code: use --print --permission-mode bypassPermissions (no PTY). Codex/Pi/OpenCode: pty:true required.
Use CodexBar CLI local cost usage to summarize per-model usage for Codex or Claude, including the current (most recent) model or a full model breakdown. Trigger when asked for model-level usage/cost data from codexbar, or when you need a scriptable per-model summary from codexbar cost JSON.
Search and analyze your own session logs (older/parent conversations) using jq.
Start voice calls via the OpenClaw voice-call plugin.
ElevenLabs text-to-speech with mac-style say UX.
Category:developer