Stats
Actions
Tags
From behavioral-core
Create a task scope before starting work. Defines boundaries, acceptance criteria, and file limits to prevent scope creep. Use before any non-trivial task.
How this skill is triggered — by the user, by Claude, or both
Slash command
/behavioral-core:scopeThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Define a focused scope for: **$ARGUMENTS**
Define a focused scope for: $ARGUMENTS
$ARGUMENTS.Glob and Grep to find relevant files — keep the list tight..claude/scopes/ if it does not exist..claude/scopes/{task-name}.md where {task-name} is a slugified version of the task (lowercase, hyphens, no spaces).The document must be under 15 lines. Use this exact structure:
# {Task Name}
## Task
{One sentence: what exactly needs to change.}
## Files
{Explicit list of files that will be touched — one per line, prefixed with `-`.}
## Boundaries
{What does NOT change — explicit exclusions, one per line, prefixed with `-`.}
## Done When
{Testable acceptance criteria — one per line, prefixed with `-`.}
## Max Files
{Number — default 5. Adjust if the user specifies a different limit.}
Present the scope document to the user and ask: "Does this scope look right? Confirm and I'll start."
Do NOT proceed with implementation until the user confirms.
npx claudepluginhub artmin96/forge-studio --plugin behavioral-coreValidates file modifications stay within authorized scope by checking user requests, task descriptions, and dependencies. Prevents scope creep and unauthorized edits.
Implements scoped code changes safely: confirms scope, edits minimal relevant files, runs checks/tests, verifies, and summarizes updates/next steps.
Creates task specifications via codebase analysis, interactive clarifying questions on scope/constraints/edge cases, feature splitting checks, and agent delegation setup.