project-scaffolding

Warn

Audited by Gen Agent Trust Hub on Oct 9, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill makes extensive use of shell commands (mkdir, touch, cp, git init, uv init, pnpm init, etc.) to initialize projects. This creates an attack surface where malicious input for project names, tech stack choices, or library names could lead to command injection if the agent does not properly sanitize these inputs before execution.
  • [PRIVILEGE_ESCALATION]: The instructions include a recommendation to use chmod +x on setup_new_project.sh to resolve permission issues. Granting execution permissions to files that may be dynamically generated or influenced by user-controlled templates is a security risk.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external project metadata which serves as a vulnerability surface for indirect prompt injection.
  • Ingestion points: Project names, descriptions, and technology stack definitions provided during the scaffolding process (SKILL.md).
  • Boundary markers: The skill lacks explicit instructions or templates that use delimiters or warnings to ignore potentially malicious instructions embedded in the project metadata.
  • Capability inventory: The skill has capabilities for file system modification (mkdir, touch, cp), package management (uv, pnpm, go, cargo), version control (git), and permission management (chmod).
  • Sanitization: There are no documented steps for sanitizing, escaping, or validating the external content before it is used in shell commands or interpolated into project configuration files like AGENT.md.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Oct 9, 2026, 01:31 AM