This is a mechanical edge case scanner that walks every branch and boundary in your code without editorializing. It's orthogonal to adversarial review: you're not looking for good or bad design, just unhandled paths. Point it at a diff and it reports only the reachable edge cases that lack explicit guards in the changed lines. Point it at a full file and it treats that as scope. The output is pure JSON with location, trigger condition, guard snippet, and consequence. No fluff, no intuition-driven hunting. Use this when you need exhaustive coverage before shipping, especially on critical paths where a missing null check or off-by-one actually matters. It won't tell you if your code is elegant, just what you forgot to handle.
npx -y skills add bmad-code-org/bmad-method --skill bmad-review-edge-case-hunter --agent claude-codeInstalls into .claude/skills of the current project.
Select a file.
prisma/skills
firebase/agent-skills
Dexploarer/hyper-forge
itsmostafa/aws-agent-skills
prisma/skills