Security risk report for @forgecat/anthropics_skills_web-artifacts-builder v0.0.5
Source Integrity
Low
Profile claims Anthropic as author with reference to official GitHub repository (github.com/anthropics/skills).
Includes specific commit hash and Apache-2.0 license, consistent with legitimate open-source attribution.
Agent Intent
Low
Content describes a legitimate web artifact builder using standard frontend technologies (React, Tailwind, shadcn/ui) with no instructions to manipulate the AI or hide/exfiltrate data.
Scripts referenced (init-artifact.sh, bundle-artifact.sh) are presented as normal development workflow steps, not as payloads or remote execution vectors.
No prompt injection, system prompt leakage, guidance poisoning, or credential-reading instructions present.
Details
Evidence
Profile describes standard React/TypeScript/Vite development workflow
Scripts are local shell commands for project initialization and bundling, not remote payloads
No instructions to read ~/.ssh, ~/.aws, .env, or exfiltrate data
Permissions
Low
No tools or MCP servers are declared in this profile, so no excessive agency is granted.
The profile is purely instructional documentation for a web artifact builder workflow.
Any shell execution would be at the user's explicit request during artifact development, not automatic or hidden.
Details
Evidence
No declared tools section
No MCP servers defined
Scripts are invoked only as part of documented development steps
MCP Risk
Low
No MCP servers are defined in this profile.
Profile contains only documentation and workflow guidance for frontend development.
ForgeCat