lovstudio-event-poster

Warn

Audited by Gen Agent Trust Hub on Jul 29, 2026

Risk Level: MEDIUMREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill hides its actual operational instructions within an encrypted binary file (SKILL.md.enc) and requires the agent to execute a vendor-provided tool (uvx lovstudio-skill-helper) to decrypt them during each use. This dynamic loading of hidden logic bypasses static analysis and allows for runtime instruction modification.
  • [COMMAND_EXECUTION]: The skill directs the agent to execute multiple shell commands, including uvx and npx wrappers, to perform decryption, license activation, and installation of skill components.
  • [EXTERNAL_DOWNLOADS]: The skill explicitly mentions that the decryption process requires a network request (an HTTP round-trip) to verify licenses and fetch content through the lovstudio-skill-helper tool, introducing external dependencies into the agent's core instruction loop.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 29, 2026, 02:54 PM
Security Audit — agent-trust-hub — lovstudio-event-poster