scope · raphaelsalaja — Claude Code skill: what it does and how to install
The scope Claude Code skill keeps code edits within the user’s request and tracks each requested item. It applies to focused changes, refactors, and planning tasks.
MIT 169 installs ★ 69 updated 2026-11-14
npx skills add raphaelsalaja/skills --skill scope -a claude-codeWhat scope does
scope tells the agent to change the named element while leaving unrelated parts alone. Before a refactor, it calls for a list of existing features and states; afterwards, the agent checks that they remain. If a request has two plausible readings, the agent asks a brief question.
The skill breaks messages into items, carries forward earlier instructions, and asks the agent to report the outcome for each item. Its checks use rule IDs and severity levels, and checks that could not run are marked Not verified. During planning, files stay untouched until the plan is approved.
What it helps with
- limit edits to the named element
- preserve working features during a refactor
- track every item in a multi-part request
- carry earlier instructions into a new edit
- keep file changes out of an unapproved plan
Example prompts for the agent
- Change only the top bar and leave the rest of the page as it is.
- Draft a refactor plan without editing files yet.
- Handle every item in my message and report what happened to each one.
How to install scope
1Skills CLI (Vercel)
npx skills add raphaelsalaja/skills --skill scope -a claude-code2Manually: put the files in place
~/.claude/skills/scope/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/scope/
1Skills CLI (Vercel)
npx skills add raphaelsalaja/skills --skill scope -a codex2Manually: put the files in place
.agents/skills/scope/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add raphaelsalaja/skills --skill scope -a cursor2Manually: put the files in place
.cursor/skills/scope/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install raphaelsalaja/skills skills/scope2Manually: put the files in place
.github/skills/scope/SKILL.md1Skills CLI (Vercel)
npx skills add raphaelsalaja/skills --skill scope -a gemini-cli2Platform CLI
gemini skills install https://github.com/raphaelsalaja/skills --path skills/scopeManually: put the files in place
.devin/skills/scope/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add raphaelsalaja/skills --skill scope -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about scope
How do I install scope in Claude Code?
Run npx skills add raphaelsalaja/skills --skill scope -a claude-code. The skill file is located at ~/.claude/skills/scope/SKILL.md.
What does scope do?
It keeps edits tied to the request, preserves earlier instructions, and tracks each requested item through the response. For refactors, it calls for checking existing features before and after the change.
What does scope require during planning?
It directs the agent to produce a plan without editing files, then wait for approval. If asked how something works, the agent should explain it and wait for another request.
Source text
SKILL.md · MIT
--- name: scope description: Changes exactly what was asked, keeps what already worked and finishes every item in the message. --- # Scope This skill holds an edit to the request. It reads the smallest literal meaning, keeps everything that already worked and accounts for every item the user sent. ## The smallest reading wins A change outside the named element is a finding unless the user asked for a sweep. One short question before a broad change is not a finding. Press hardest on short messages that point at one element, on batched messages and on refactors. Deleting what nobody asked for belongs to `restraint`. Proving the change works belongs to `verify`. ## Every ask gets an answer Split the message into numbered items and add the open items from earlier turns. Close with the same list, each item done or not done with the reason. Doing the first of four asks is `scope-list-every-ask`. ## The named element and nothing around it Restate the target in a few words. "The card" means the card, not the page around it. Restyling the whole shell for a top bar request is `scope-smallest-reading`. When two readings fit, ask one short question. ## Earlier instructions still stand Before editing, read the user's earlier corrections in the thread and the rules in the repository. Re-adding something they removed is `scope-remember-instructions`. A value they gave, such as `16px` of left padding, stays. Guessing a new one is `scope-keep-given-values`. ## What worked keeps working Before a refactor, list the features, buttons and states that exist now, then check each one after. A dropped back button is `scope-keep-what-worked`. When fixing one property, keep the others. Fixing a clip by removing the width animation is `scope-no-collateral`. ## Unshipped work is done now Work that has not shipped is not split into phases or marked out of scope without asking. A "phase 2" list is `scope-no-phasing`. While planning, plan only. Editing files before the plan is approved is `scope-plan-means-plan`. ## An explanation is the answer When the user asks how something works, explain it and wait. Answering "how does this work" with a diff is `scope-explain-when-asked`. ## The pair Every check is a pair. The broken line sits beside the line that replaces it. | Rule | Broken | Replacement | | --- | --- | --- | | `scope-list-every-ask` | The first of four asks done, three dropped | Four numbered items, each reported | | `scope-smallest-reading` | "Fix the top bar" restyles the whole shell | Only the top bar changes | | `scope-remember-instructions` | Dots the user removed come back | Earlier turns read before editing | | `scope-keep-given-values` | Left padding re-guessed as `12px` | The `16px` the user gave | | `scope-keep-what-worked` | A refactor drops back and forward buttons | Before and after feature list, both buttons kept | | `scope-no-collateral` | The clip is fixed and the width animation is gone | Clip fixed, width animation kept | | `scope-no-phasing` | "Saved filters: phase 2" | Built now, or one question | | `scope-plan-means-plan` | Files edited during planning | Plan only until it is approved | | `scope-explain-when-asked` | "Explain how it works" answered with a diff | An explanation, then wait | ## Reporting **Severity.** `HIGH` is a dropped feature, or an earlier instruction undone. `MEDIUM` is a change outside the named element, or an ask left unreported. `LOW` is a question asked where one reading was plain. **Format.** One finding per root cause, citing the request when no file applies: ``` file:line - [rule-id] what is wrong request 2 - [rule-id] what is wrong ``` Then a `Rule | Count | Severity` summary. `Block` when any `HIGH` remains, `Approve` otherwise. Never `Approve` a diff you did not read. Mark every check you could not run as `Not verified`.
© raphaelsalaja · MIT · raphaelsalaja/skills
Similar skills
SKILL.md skills catalog →scope by other authors comes first.