Skip to content

Scope Check

Compares a feature or sprint's original plan against current work to flag and quantify scope creep.

Stars
32,965
Licence
MIT
Files
1 file

What it does

This skill makes the agent act as a scope auditor for a project. You give it a feature, sprint, or milestone name; it finds the original plan document, looks at what has actually been built or committed (via code search and git history), and produces a report listing added and removed items, a percentage change, risk ratings, and suggested cuts or deferrals. It is read-only and writes no files, and it returns a NOT ASSESSED verdict instead of a misleading 0% when the plan or current state can't be read.

Great for

  • Game or software developers running sprints who suspect the work has grown beyond the plan
  • Producers or leads who want a quantified, evidence-based view of scope before a milestone
  • Students or small teams using a structured design-doc workflow

Skip it if

  • Projects that do not keep a written plan or design document, since the skill has nothing to compare against and will report NOT ASSESSED
  • Anyone expecting a precise objective measurement: item counts and the verdict depend on how the plan and the agent interpret the documents, so results are advisory

Try asking

  • “/scope-check user-login-flow”
  • “/scope-check sprint-3”
  • “/scope-check combat-system milestone review”
  • “Any scope creep on the inventory feature?”
Step-by-step guide: how to use Scope CheckInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 1 original files, the MIT licence and an attribution note.

    scope-check.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip scope-check.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip scope-check.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Keep the baseline plan document up to date with a clear list of scope items, otherwise the verdict will be NOT ASSESSED.
  • Re-run the check after making cuts to confirm the verdict improves.
  • Treat the percentages as a starting point for a discussion with the team, not a final judgement.

What's inside

  • SKILL.md

Redistributed unmodified from Donchitos/Claude-Code-Game-Studios under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.