new-element

Pass

Audited by Gen Agent Trust Hub on Jul 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands ('cp', 'mv', 'bun install') that incorporate placeholders such as '' and ''.
  • Evidence: Bash blocks in sections 2 and 4 use these placeholders to construct file paths and move files within the 'packages/docs' directory.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes user-provided inputs ('category', 'slug') and interpolates them into shell commands without explicit sanitization or boundary markers.
  • Ingestion points: User-supplied placeholders for '' and '' in SKILL.md.
  • Boundary markers: Absent; placeholders are placed directly into shell command strings.
  • Capability inventory: File system operations ('cp', 'mv'), code formatting ('bunx oxfmt'), and package installation ('bun install') via shell execution.
  • Sanitization: Absent; the skill does not instruct the agent to validate or escape the provided strings before use in the shell.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 16, 2026, 12:28 PM
Security Audit — agent-trust-hub — new-element