boilerplate

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill relies on shell scripts (bash) to perform framework detection and template substitution. It uses commands like grep, sed, cat, and source.
  • Evidence: Use of sed "s/\${COMPONENT_NAME}/$component_name/g" and source "$cache_file" in the optimization logic described in SKILL.md.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection or command injection because it interpolates user-provided arguments ($ARGUMENTS) directly into shell commands without sanitization.
  • Ingestion points: User-supplied arguments for component/route/model names (SKILL.md).
  • Boundary markers: None present to distinguish user input from executable command logic.
  • Capability inventory: File system read/write, shell command execution via agent capabilities.
  • Sanitization: No evidence of validation or escaping for the $ARGUMENTS variable before its use in sed or file path construction.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 05:05 PM
Security Audit — agent-trust-hub — boilerplate