claude-skill-surgical-changes

claude-skill-surgical-changes is a skill for Claude Code from ishandutta2007/claude-skills-by-karpathy. It costs 34 tokens per session (211 once invoked), scanned A, original, MIT.

A set of rules for making small, focused code edits that match the existing project style.

In plain words
What is it for?
Use it when editing an existing codebase to limit changes to what is necessary, clean up only new unused code, and report older dead code without removing it.
Why use it?
It reduces unrelated changes and avoids refactoring code that was not part of the request.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the claude-skills-by-karpathy plugin — 4 skills shipped together

Good fit Use it when editing an existing codebase to limit changes to what is necessary, clean up only new unused code, and report older dead code without removing it.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/ishandutta2007/claude-skills-by-karpathy/surgical-changes
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add ishandutta2007/claude-skills-by-karpathy --skill surgical-changes
Clone the repo
git clone --depth 1 https://github.com/ishandutta2007/claude-skills-by-karpathy

Made for: Claude Code.

Or install claude-skills-by-karpathy, the plugin that ships this one along with the rest of its 4 skills.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for claude-skill-surgical-changes

README.md
[![agentmods](https://agentmods.dev/badge/skills/ishandutta2007/claude-skills-by-karpathy/surgical-changes/github.svg)](https://agentmods.dev/skills/ishandutta2007/claude-skills-by-karpathy/surgical-changes)
Your own site
<a href="https://agentmods.dev/skills/ishandutta2007/claude-skills-by-karpathy/surgical-changes"><img src="https://agentmods.dev/badge/skills/ishandutta2007/claude-skills-by-karpathy/surgical-changes/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for claude-skill-surgical-changes

Your own site · 80×15
<a href="https://agentmods.dev/skills/ishandutta2007/claude-skills-by-karpathy/surgical-changes"><img src="https://agentmods.dev/badge/skills/ishandutta2007/claude-skills-by-karpathy/surgical-changes.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 34 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 211 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00034 $0.00211
Opus 5 $0.00017 $0.00105
Sonnet 5 $0.00007 $0.00042
Haiku 4.5 $0.00003 $0.00021

Measured 12d ago against content hash 8c8c4336a138, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

claude-skill-surgical-changes scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

claude-skills/surgical-changes/SKILL.md · 22 lines

What it actually says

Surgical Changes (Claude Skill)

Touch only what you must. Clean up only your own mess.

When editing existing code:

  • Don't "improve" adjacent code. Avoid touching comments or formatting unrelated to the task.
  • Don't refactor what isn't broken. Keep the diff clean and focused.
  • Match existing style. Even if you prefer a different pattern, follow the project's lead.
  • Report unrelated dead code. Mention it in the output instead of deleting it unless asked.

When your changes create orphans:

  • Clean up your own mess. Remove imports, variables, or functions that YOUR changes made unused.
  • Leave pre-existing dead code alone. Unless explicitly requested to clean it up.

The test: Every changed line should trace directly to the user's request.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 12d ago First seen · 22 lines · 34 tokens per session scan A 8c8c4336a138

Subscribe to this mod's changes

claude-skill-surgical-changes is a skill published in the GitHub repository ishandutta2007/claude-skills-by-karpathy (7 stars, last pushed 12d ago), licensed MIT. It adds 34 tokens to every session and 211 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

prompt-master

Generates optimized prompts for AI tools. Activates only when the user explicitly asks to write, fix, improve, or adapt a prompt for a specific AI tool (LLM, Cursor, Midjourney, image AI, video AI, coding agents, etc.). Does not activate for general conversation, coding tasks, document writing, or other…

nidhinjs/prompt-master · 78 tokens

api-tester

A tool for creating and checking API tests from the real API contract and implementation. An API is the agreed way that software sends requests and receives responses.

laolaoshiren/claude-code-skills-zh · 86 tokens

dep-auditor

A read-only audit of a project’s third-party packages for known security issues, version health, and license facts. It uses the versions actually selected by lockfiles, which record the exact packages installed, rather than relying only on version ranges.

laolaoshiren/claude-code-skills-zh · 72 tokens

skill-curator

A Chinese-language evaluator for deciding whether developer tools and agent resources are suitable for a curated collection. It checks real repositories, installation paths, activity, duplicates, and security boundaries using evidence.

laolaoshiren/claude-code-skills-zh · 75 tokens

git-workflow

A guide for handling Git repository work safely, including status checks, branches, commits, pushes, pull requests, and rebasing. Git is a version-control system that records code changes and coordinates work between developers.

laolaoshiren/claude-code-skills-zh · 73 tokens

zh-readme

A tool for writing README files in Chinese after first examining a software project. A README is the main guide visitors see when they open a code repository.

laolaoshiren/claude-code-skills-zh · 25 tokens