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.
npx agentmods add skills/orlando-japan/claude-code-setting/surgical-changesnpx skills add orlando-japan/claude-code-setting --skill surgical-changesgit clone --depth 1 https://github.com/orlando-japan/claude-code-settingWrote 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.
[](https://agentmods.dev/skills/orlando-japan/claude-code-setting/surgical-changes)<a href="https://agentmods.dev/skills/orlando-japan/claude-code-setting/surgical-changes"><img src="https://agentmods.dev/badge/skills/orlando-japan/claude-code-setting/surgical-changes.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00036 | $0.00560 |
| Opus 5 | $0.00018 | $0.00280 |
| Sonnet 5 | $0.00007 | $0.00112 |
| Haiku 4.5 | $0.00004 | $0.00056 |
Grade A, and why
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 6d 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.
How it starts
The opening of the file, as written. The whole thing — 54 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Surgical changes
The goal of a surgical change: the diff is small, focused, and obviously correct. Reviewers shouldn't have to wonder which lines are the "real" change.
Apply when
- Fixing a bug.
- Making a targeted feature addition.
- Any change that will be reviewed (which is all of them).
Rules
- Touch only lines the task requires. If you find yourself about to edit a file the task doesn't mention, stop and justify.
- Don't reformat. If the file uses 2-space indent and you prefer 4, use 2. If it uses
letand you preferconst, uselet. Match what's there. - Don't rename unrelated things. A rename is its own commit.
- Don't re-order. Reordering declarations, imports, or methods for aesthetic reasons adds noise to the diff.
- Don't add unrelated tests. Fix your bug with its regression test. That's it.
- Don't remove unrelated dead code. Put it on a todo list; do it separately.
- If you notice a real bug unrelated to the task, stop and tell the user. Don't silently fix it in the same change.
Before committing
Look at the diff. For every hunk, ask: "Does this line directly implement the task?" If the answer is no or "sort of," revert that hunk.
Typical cuts:
- Whitespace-only changes → revert.
- Import re-sorting → revert.
- Comment rewording → revert unless the comment was actively wrong.
- Variable renames → revert unless ambiguity was causing the bug.
- "Cleanup" of unrelated logic → revert and file a separate ticket.
Exception — fixing on the way through
Rarely, small unrelated fixes inside a file you're already touching are acceptable:
- Obvious typo in a string literal that will ship in UI.
- Misleading comment right next to your change that would confuse the reviewer.
console.logleft over from someone else's debugging in code you're editing.
Call these out explicitly in the commit body: "Also fixed X — unrelated but one-line."
Why this matters
Small diffs are reviewable. A review that finds a real bug is worth more than a diff that silently improves ten things. Bundled cleanup also attributes blame wrong when git blame runs.
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.
- 6d ago First seen · 54 lines · 36 tokens per session scan A 0e898705c986
surgical-changes is a skill published in the GitHub repository orlando-japan/claude-code-setting (2 stars, last pushed 3mo ago), licensed MIT. It adds 36 tokens to every session and 560 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.
Other skills, from other repositories
claude_code_delegation
Delegate coding, review, and refactor tasks to Claude Code CLI through Row-Bot's approval-gated shell workflow.
military-discipline
严格执行，质量把关，拒绝敷衍.
pr-review
Review a GitHub pull request and post one formal review — advance the existing discussion and give precision-first, high-signal feedback. Judgement on the diff, not a build gate — CI validates that it builds, and a targeted probe is allowed as evidence. Use when asked to review a PR or on a cron PR scan.
code-review
The shared rubric for reviewing a code change — precision-first, high-signal findings only (Critical/Important), what to check (correctness, contracts between producer and consumer, security, edge cases, regression risk, test coverage), and what is NOT a finding (pre-existing issues, linter territory, aspirational…
Code Review
Review a patch, diff, PR, or implementation for bugs, regressions, risky assumptions, and missing tests.
adjudicate-pass
The shared rules for the ADJUDICATE pass of a multi-pass PR review — the workspace layout, the severity vocabulary it ranks on, and the prior-review ledger plus the two APPROVE constraints it inherits as the only pass that owns the posted verdict. Use when turning a pile of hypotheses into one ranked, tiered review.…