ask-questions-if-underspecified

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill consists entirely of instructional markdown text and does not include scripts, binary files, or automated command execution.
  • [INDIRECT_PROMPT_INJECTION]: The skill describes a surface for processing untrusted data. 1. Ingestion points: User requests and repository configuration files. 2. Boundary markers: Not specified within the skill instructions. 3. Capability inventory: Limited to inspecting repository structure and reading configuration files. 4. Sanitization: No explicit content validation or sanitization is instructed.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 11:36 PM
Security Audit — agent-trust-hub — ask-questions-if-underspecified