T2I-Studio-Premium
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: HIGHEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- EXTERNAL_DOWNLOADS (HIGH): The skill's installation instructions (npm install -g CatfishW/T2IAgentSkill) point to a personal GitHub repository rather than a verified organization or a standard registry. This bypasses security vetting and allows the author to push malicious updates directly to the user's system.
- COMMAND_EXECUTION (HIGH): The core functionality in
bin/t2i.jsuseschild_process.spawnto execute a Python script located at../client.py. Since this script is part of the untrusted remote package and was not provided for auditing, it constitutes unverified code execution at runtime. - Metadata Poisoning (MEDIUM): The
SKILL.mdclaims the package uses 'Private distribution via authenticated registry' and is 'Secure'. These claims are demonstrably false as the installation method uses a public, unauthenticated GitHub repository, which could mislead users into a false sense of security.
Recommendations
- AI detected serious security threats
Audit Metadata