checklist-runner

Pass

Audited by Gen Agent Trust Hub on Mar 16, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill processes local markdown files and instructs the agent to follow embedded instructions, which constitutes an indirect prompt injection surface. This is considered safe as checklists are sourced from a local, developer-controlled directory (.aiox-core/development/checklists/).
  • Ingestion points: .aiox-core/development/checklists/{name}.md.
  • Boundary markers: Absent; no specific instructions provided to ignore or delimit adversarial content.
  • Capability inventory: The skill gathers context by reading documentation, source code, and outputs from local development tools (npm test, git diff).
  • Sanitization: Absent; the engine does not validate or sanitize instructions found within the checklists.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 16, 2026, 11:45 PM