# Scope Sweep

> Final pass to identify missed items, edge cases, and risks before considering a scope done. Use as a final breadth pass before calling a scope or task done - to catch missed items, overlooked requirements, edge cases, and risks after the core work is complete and reviewed.

- Skill: `mahdtech/scope-sweep` (Agent Skill)
- Install (CLI): `npx skillmds@latest add mahdtech/scope-sweep`
- Raw SKILL.md: https://api.skillmd.com/api/skills/mahdtech/scope-sweep/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Productivity
- Author: MAHDTech (https://skillmd.com/u/mahdtech)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/mahdtech/scope-sweep

---


# Scope Sweep

Now that we've completed the core work and reviewed it, do a final pass to identify anything we missed or should address before considering this scope "done".

This is the **breadth pass** - it makes sure nothing was dropped. For a deep dive on a single axis, reach for the focused skill instead: `/pr-edge-cases` for logic and failure-mode analysis of a diff, `/self-review` for simplification and idiomatic cleanup, `/critical-thinking` to stress-test your own reasoning.

_Note: Prefer your agent's built-in file-search and file-reading tools (instead of ad-hoc bash `find`/`grep`) to quickly sweep the codebase for incomplete items, FIXME comments, or test gaps._

Please check for:

- **Incomplete tasks**: anything you planned to do but didn't get to
- **Overlooked requirements**: anything implied but not explicitly handled
- **Missing questions**: what I should have asked but didn't
- **Hidden assumptions**: any assumptions you made that could be wrong
- **Edge cases**: scenarios that could break or degrade the solution
- **Failure modes**: what happens when inputs are malformed, missing, hostile, or extreme
- **Operational concerns**: deployment, monitoring, logging, maintenance, rollback
- **Security and abuse risks**: permissions, misuse paths, data exposure
- **Performance bottlenecks**: scalability issues, worst-case behavior, cost traps
- **Testing gaps**: what still needs tests or validation
- **User experience gaps**: confusing flows, unclear outputs, poor defaults
- **Integration risks**: dependencies, external APIs, versioning, backward compatibility
- **Future-proofing**: what will become painful in 3-6 months if not handled now

Finally:

- **What is the biggest risk remaining?**
- **If you were reviewing this as a third party, what would you criticize first?**
- **What would you improve if you had one more day?**