skills/openlark/skills/perryts/Gen Agent Trust Hub

perryts

Pass

Audited by Gen Agent Trust Hub on Aug 8, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides instructional content for a legitimate developer toolchain, PerryTS, without any malicious directives.
  • [EXTERNAL_DOWNLOADS]: Documents standard installation procedures for the @perryts/perry compiler and its dependencies through official package managers like npm, Homebrew, and Winget.
  • [COMMAND_EXECUTION]: Lists valid development commands for compiling code, managing projects, and setting up build toolchains, all consistent with the skill's stated purpose.
  • [SAFE]: Describes a local UI testing server ('Geisterhand') that runs on the loopback interface (127.0.0.1:7676) for automation, posing no risk of remote data exfiltration.
  • [SAFE]: Explicitly defines architectural limitations that prevent common attack vectors, such as the lack of support for eval() or dynamic prototype mutation.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 8, 2026, 04:17 AM
Security Audit — agent-trust-hub — perryts