Turn user requests into atomic, actionable development plans

Concise Planning turns a user request into a single, actionable plan made of atomic steps. It runs a three-step workflow: scan the repo's README.md, docs, and relevant code to identify constraints such as language, frameworks, and tests; ask at most 1-2 questions, only if something is truly blocking, making reasonable assumptions otherwise; then generate a plan using a fixed structure - a 1-3 sentence Approach, an In/Out Scope, 6-10 atomic verb-first Action Items in order, and at least one Validation item for testing. The output follows a markdown template with checkbox action items and an Open Questions section capped at 3 questions.

Use it when a request needs to be turned into a concrete, ordered plan before execution - discovery through rollout - rather than jumping straight into changes. Skip it, or stop and ask for clarification, when required inputs, permissions, safety boundaries, or success criteria are missing; the skill is not a substitute for environment-specific validation, testing, or expert review of the resulting plan.

Input: a user request plus the target repository's README.md, docs, and relevant code files. Output: a markdown plan with Approach, Scope (In/Out), 6-10 atomic Action Items (Add, Refactor, Verify, and similar verb-first steps naming specific files or modules where possible), at least one Validation/Testing item, and an Open Questions list of up to 3 items.

• Reads the project's own README.md, docs, and code as its only context source
• Produces a checkbox-style markdown plan meant to be handed off to implementation

Anyone using an AI agent to turn a request into a scoped, step-by-step plan before code changes begin, when they want a consistent plan format rather than free-form prose.