lovstudio-xbti-gallery

Pass

Audited by Gen Agent Trust Hub on Jul 29, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The skill interacts with external services by opening a browser to the author's domain (xbti.lovstudio.ai) and using the gh CLI to fetch information from GitHub. These are well-known or vendor-controlled services used for their intended purpose.
  • [DYNAMIC_EXECUTION]: The skill executes an inline Python script using python3 -c to parse and format JSON data retrieved from the GitHub API. This is a static script used for local data processing.
  • [INDIRECT_PROMPT_INJECTION]: The skill reads directory names from a public GitHub repository and displays them to the user. While this creates a potential surface for indirect injection if the repository content were malicious, the source is the author's own repository and the output is presented as a list of items.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 29, 2026, 08:00 AM
Security Audit — agent-trust-hub — lovstudio-xbti-gallery