scaffold-exercises

Pass

Audited by Gen Agent Trust Hub on Jul 11, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands such as mkdir -p, git mv, and git commit to manage the exercise file system structure.
  • [COMMAND_EXECUTION]: It invokes a local command pnpm ai-hero-cli internal lint to perform automated validation of the created stubs.
  • [PROMPT_INJECTION]: The skill contains an indirect prompt injection surface where user-provided exercise plans are used to construct directory paths and file content. Evidence Chain: (1) Ingestion points: Plan parsing in Workflow step 1; (2) Boundary markers: Absent; (3) Capability inventory: Subprocess calls (mkdir, pnpm) and file writes in SKILL.md; (4) Sanitization: Absent. This surface is considered low-risk given the skill's intended development purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 11, 2026, 12:52 PM
Security Audit — agent-trust-hub — scaffold-exercises