grill-me

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill consists entirely of markdown-based natural language instructions. There are no scripts, binaries, or automated shell commands present in the skill definition.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted user input (plans and designs) and has the capability to explore the local codebase. This represents a standard surface for indirect prompt injection, although the risk is minimal given the lack of executable actions defined within the skill. Evidence Chain: (1) Ingestion Points: User plans/designs provided during the conversation. (2) Boundary Markers: Absent. (3) Capability Inventory: Codebase exploration (read-only file access). (4) Sanitization: Absent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 10:05 AM