skills/simota/agent-skills/Gear/Gen Agent Trust Hub

Gear

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: References installation scripts and GitHub Actions from trusted and well-known providers like bun.sh, Astral, and Gitleaks.
  • [COMMAND_EXECUTION]: Executes shell commands via package managers (pnpm, uv, cargo) and Docker to manage project builds and environments.
  • [PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection as it processes untrusted project data.
  • Ingestion points: Reads package.json, Dockerfiles, and CI workflows.
  • Boundary markers: Mentions the use of _common/BOUNDARIES.md to define role constraints.
  • Capability inventory: Capable of file writing and command execution using pnpm and Docker CLI across reference files.
  • Sanitization: No explicit input sanitization or validation of project file content is documented in the references.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 09:07 PM