awesome-gamedev-agent-skills

Warn

Audited by Gen Agent Trust Hub on Oct 29, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to use npx -y skills@1.7.0 to fetch a package from the npm registry, which is an external download of executable code.
  • [REMOTE_CODE_EXECUTION]: The use of npx to execute the skills package constitutes remote code execution. The -y flag is used to skip the installation prompt, which reduces user oversight of the download process.
  • [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes SKILL.md files from an external repository (gamedev-skills/awesome-gamedev-agent-skills). * Ingestion points: Markdown instruction files are read from a user-supplied local directory containing content from the external repository as described in SKILL.md. * Boundary markers: The skill instructs the agent to manually review content ('read each selected leaf body before adopting it'), but no programmatic boundaries are defined to isolate the external instructions. * Capability inventory: The skill has the capability to perform file writes via the skills tool and command execution via the Bash tool, both permitted in SKILL.md. * Sanitization: No evidence of automated sanitization or escaping of the external content is present.
  • [COMMAND_EXECUTION]: The skill executes a bundled Python script (scripts/preflight.py) for integrity checks and uses the npx command for tool execution.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Oct 29, 2026, 09:45 AM