Skip to content

softaworks/agent-toolkit / skills/reducing-entropy

reducing-entropy

Unverified

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.

adding-reference-mindsets.md

View on GitHub
# Adding Reference Mindsets

Reference mindsets are philosophical foundations for simplicity. They explain *why* less is more, giving agents deeper calibration beyond the mechanical checkpoints.

## Where They Live

Mindsets live in @reference/. Each is a standalone file named by concept.

## File Structure

```yaml
---
description: One-sentence summary of the core insight and why it matters.
---

# Concept Name

## The Core Insight

The central idea in 1-2 sentences. Quotable. Memorable.

## Why This Matters

How this connects to avoiding complexity. Why an LLM should care.
What goes wrong when you ignore this principle.

## Practical Application

Concrete questions to ask or checks to apply.
How to use this mindset when evaluating design options.

## External References

Links to primary sources - talks, papers, books that originated or best explain this concept.
```

## Quality Checklist

Before adding a mindset:

- [ ] **Counters over-engineering?** Does it help resist the urge to add?
- [ ] **Distinct from existing?** Not redundant with current mindsets
- [ ] **Concise?** Can be explained in under 50 lines
- [ ] **Memorable core insight?** Has a quotable central principle
- [ ] **Named by concept?** Not by person or source

## Good Candidates

Ideas that would make strong mindsets:

| Concept | Core Insight |
|---------|--------------|
| `worse-is-better` | Shipping a simple thing beats perfecting a complex one |
| `essential-vs-accidental` | Most complexity is accidental and can be eliminated |
| `locality-of-behavior` | Code should be understandable without jumping around |
| `boring-technology` | Innovation tokens are limited; use boring tech by default |
| `separation-of-concerns` | Each piece should have one reason to change |
| `rule-of-three` | Don't abstract until you've seen the pattern three times |

## What NOT to Add

**Technology-specific advice** → Belongs in project docs or tech-specific skills
- "React components should..." 
- "In Rust, prefer..."

**Process/workflow rules** → Belongs in skills, not mindsets
- "Always run tests before..."
- "Use TDD when..."

**Vague platitudes** → If there's no actionable insight, skip it
- "Write clean code"
- "Think before you code"

**Anything requiring context** → Mindsets should be universal
- "In microservices architectures..."
- "When working with legacy code..."

## The Test

A good mindset should help an agent answer: "Should I add this abstraction?"

If the mindset doesn't directly inform that question, it probably belongs somewhere else.

## Source Material

Primary sources for simplicity thinking. Draw from these when creating new mindsets:

- [Simple Made Easy](https://www.infoq.com/presentations/Simple-Made-Easy/) - Rich Hickey
- [Hammock Driven Development](https://www.youtube.com/watch?v=f84n5oFoZBc) - Rich Hickey
- [The Value of Values](https://www.infoq.com/presentations/Value-Values/) - Rich Hickey
- [Out of the Tar Pit](https://curtclifton.net/papers/MosesleyMarks06a.pdf) - Moseley & Marks
- [No Silver Bullet](https://www.cs.unc.edu/techreports/86-020.pdf) - Fred Brooks
- [The Grug Brained Developer](https://grugbrain.dev/) - grugbrain.dev
- [Worse Is Better](https://www.dreamsongs.com/WorseIsBetter.html) - Richard Gabriel
- [A Philosophy of Software Design](https://www.amazon.com/dp/173210221X) - John Ousterhout
- [The Zen of Python](https://peps.python.org/pep-0020/) - Tim Peters
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 463kUpdated Claude CodeCodexVerifiedInstall
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 463kUpdated Claude CodeCodexVerifiedInstall
discernment-nudgeanthropics/skillsAfter you give a substantive answer or draft that the user may act on — advice or recommendations, drafted artifacts such as goals, plans, pitches, proposals, or emails, estimates or projections, analysis or interpretation of data, factual claims they may rely on, or a multi-step argument — invoke this skill BEFORE finalizing your reply and then, if it applies, append 2-3 short follow-up questions, each tied to something specific in what you just produced, that help the user check key facts, probe the reasoning or assumptions, and notice missing context. Do this at most once per conversation. Skip it when the user asked a trivial how-to or simple lookup, wants a purely educational explanation, asked you only to format, convert, or assemble a file from content they provided, is writing code they will run, is doing creative writing or casual chat, or already asked you to double-check, cite, or review — the skill file explains these boundaries and the exact output format.Stars: 195kUpdated Claude CodeCodexVerifiedInstall
internal-commsanthropics/skillsA set of resources to help me write all kinds of internal communications, using the formats that my company likes to use. Claude should use this skill whenever asked to write some sort of internal communications (status reports, leadership updates, 3P updates, company newsletters, FAQs, incident reports, project updates, etc.).Stars: 195kUpdated Claude CodeCodexVerifiedInstall