How do I install the Swift Review skill?

Run npx skillmds@latest add swiftzilla/swift-review in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Swift Review safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Swift Review?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Swift Review free to use?

Yes. Installing skills from SkillMD is free. This skill is licensed under Proprietary.

Who published Swift Review?

swiftzilla (@swiftzilla) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Swift Review by swiftzilla — Agent Skill

Name: Swift Review
Author: swiftzilla

Swift Review

Use this skill for PR safety and risk analysis. It analyzes the risk and blast radius of a PR using git diff. Use before committing changes, reviewing pull requests, or assessing the impact of code modifications — even if they don't mention "review" but describe checking change safety or understanding what a diff affects.

swiftzilla b48215b 3 files · 27.7 MB Updated Sep 17, 2026

File contents

Swift Review: PR Safety & Risk Analysis

Analyze the semantic impact of your changes and catch potential regressions before they land.

Commands

# Review diff against current HEAD
scripts/swift_review

# Review diff against a specific target branch
scripts/swift_review --target <BRANCH>

Workflow

Analyze: Run scripts/swift_review to see the semantic impact of your changes
Assess risk: Look for "High Impact" warnings (high Fan-In files) — these are the most dangerous changes to get wrong
Verify: For each file mentioned in the "Impacted Files" list, run its specific tests to confirm nothing broke
Proceed: Only commit when high-risk areas are covered by tests

Gotchas

Always run this BEFORE committing — not after. The tool analyzes uncommitted changes by default.
"High Impact" means the file has many dependents (high Fan-In). Changes to these files have a larger blast radius and need more careful testing.
If no tests exist for an impacted file, flag it for manual review — the tool cannot verify correctness without tests.
When using --target, make sure the target branch is up to date locally — stale branches produce misleading diffs.

Frequently asked questions