skill-bootstrap

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local shell commands to run a validation script (bash tools/validate.sh) and to clean up temporary scaffold files (rm -f). These operations are confined to the local repository context.
  • [COMMAND_EXECUTION]: The skill uses a user-defined skill name to construct shell commands and file paths. Although the instructions recommend a kebab-case format, the logic relies on the agent to correctly format the input before execution.
  • [SAFE]: The skill promotes security best practices for AI development, such as Test-Driven Development (TDD), the inclusion of adversarial test prompts to check rule enforcement, and a 'Red Flags' table to prevent agents from rationalizing around safety constraints.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 04:35 PM
Security Audit — agent-trust-hub — skill-bootstrap