reducing-entropy_skill

This skill helps you minimize the total codebase by prioritizing deletion and measuring final size, ensuring the smallest possible solution.
  • Python

273

GitHub Stars

3

Bundled Files

3 weeks ago

Catalog Refreshed

2 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstart where the catalogue uses aiagentskills.

npx veilstart add skill softaworks/agent-toolkit --skill reducing-entropy

  • adding-reference-mindsets.md3.3 KB
  • README.md4.7 KB
  • SKILL.md2.5 KB

Overview

This skill is a manual-only workflow for minimizing total codebase size with a strong bias toward deletion. It prioritizes the final amount of code as the success metric and only runs when explicitly requested. Use it when the goal is a smaller, simpler codebase rather than incremental improvements.

How this skill works

Before making any edits, you load at least one mindset document from references/ and state which mindset and principle you adopted. You evaluate proposed changes by asking three core questions: what is the smallest final codebase that still solves the problem, will the change reduce total lines of code, and what can be deleted as part of the change. Count lines before and after; accept only changes that reduce total code.

When to use it

  • When the primary objective is to minimize total code size rather than feature expansion.
  • During refactors aimed at consolidation or removal of redundant functionality.
  • Before merging large feature branches to decide if features should be removed or simplified.
  • When technical debt manifests as excess files, duplicate logic, or unnecessary abstractions.
  • When preparing a repository for archival, handoff, or a constrained deployment target.

Best practices

  • Always load and state at least one reference mindset from references/ before changing code.
  • Measure net lines of code before and after each candidate change; reject increases.
  • Favor deletion of whole features or files when they are not essential; deleting 200 lines to add 50 is a win.
  • Question abstractions that add files or functions; simplicity with fewer lines is preferred.
  • Document what was removed and why, referencing the mindset used to justify deletion.

Example use cases

  • Remove a deprecated feature and all its tests and docs, keeping only the minimal surviving API.
  • Consolidate multiple tiny utility modules into one small helper file, deleting the originals.
  • Replace a complex plugin system with a single hardcoded implementation where flexibility is unused.
  • Eliminate generated code paths that are never invoked and the supporting scaffolding.
  • Prune platform-specific branches and fallbacks when target platforms have converged.

FAQ

No. This is a manual workflow and decision framework. It guides what to delete and how to measure success but does not modify files automatically.

What if deleting breaks tests or users?

Prioritize end-state functionality. If tests or users depend on behavior, either preserve minimum required code or remove the feature and update stakeholders and tests accordingly.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational