scaffold-exercises

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands including mkdir -p for directory creation, git mv for file management, and pnpm ai-hero-cli internal lint for validation. These commands are standard for project scaffolding and use local development tools.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through its ingestion of untrusted data.
  • Ingestion points: The agent is instructed to parse a user-provided "plan" to extract directory and exercise names (Workflow Step 1).
  • Boundary markers: There are no instructions for the agent to use delimiters or ignore embedded instructions within the user's plan.
  • Capability inventory: The agent has access to file system modification tools (mkdir, git mv) and the project's linter via pnpm.
  • Sanitization: There are no explicit sanitization or validation steps for the inputs parsed from the external plan before they are used in directory naming or shell commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 10:16 AM
Security Audit — agent-trust-hub — scaffold-exercises