skills/k5-mot/skills/init-project/Gen Agent Trust Hub

init-project

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill automates the installation of external agent skills using pnpm dlx skills@latest. These downloads target identified user repositories (k5-mot/skills, mattpocock/skills) and utilize the standard pnpm ecosystem for package acquisition. This is a common pattern for environment setup skills and is considered safe when targeting known community providers.
  • [COMMAND_EXECUTION]: The skill uses uv run and pnpm exec to initialize project tools (graphify and openspec). These are legitimate development tools for repository architecture and specification management. While they involve local command execution, the operations are restricted to project initialization and configuration management within the current directory.
  • [SAFE]: No signs of prompt injection, obfuscation, or suspicious persistence mechanisms were detected. The use of --force in the OpenSpec initialization is explicitly documented with a warning about potential configuration overwrites, following best practices for user transparency.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 04:49 PM
Security Audit — agent-trust-hub — init-project