Skip to content
Back to skills

Blast Radius

ASecurity

Find what a Roblox change can break across client/server contracts, remotes, saved data, replication, authored Instances, mobile performance, and dependent systems.

  • 2 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added September 19, 2026
ai-agentstestingsecurityperformance

Works with

  • cli

Security analysis

A100/100

Pro scans all 2 files and shows the line behind each finding

Scanned September 19, 2026

npx -y skills add HungryKelvin123/dstack --skill blast-radius --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Blast Radius?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Blast Radius
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/hungrykelvin123-blast-radius/badge)](https://www.skillsdirectory.com/skills/hungrykelvin123-blast-radius)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: blast-radius
description: "Find what a Roblox change can break across client/server contracts, remotes, saved data, replication, authored Instances, mobile performance, and dependent systems."
---

# Roblox blast-radius review

Inspect the diff, then follow consumers beyond it.

- Trace callers, callees, required modules, remotes, attributes, tags, CollectionService consumers, project mappings, and authored Instance paths.
- Check whether a client/server payload, replicated shape, saved schema, migration, reward identity, cooldown, or lifecycle assumption changed.
- Check reconnects, duplicate requests, player removal, character replacement, round reset, server shutdown, and retry behavior.
- Check mobile CPU, memory, physics, render, network, and instance-count costs.
- Check security boundaries. A client-visible or client-supplied value must not become authority by accident.
- Run the smallest check that proves each material risk. Use source and local automation. Studio playtesting belongs to the user.

Report only concrete risks. Each finding names the failure, evidence path, likelihood, impact, and verification. End with the strongest fact supporting safety and any remaining Studio test.

Files in this skill

  • SKILL.md1.2 KB
  • agents/openai.yaml206 B

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.