qa-testcase-from-docs

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process requirements and specification documents from external sources, creating a surface for indirect prompt injection where hidden instructions in the source text could attempt to influence agent behavior.\n
  • Ingestion points: Requirements and specification documents provided as input to the workflow in SKILL.md.\n
  • Boundary markers: There are no explicit delimiters or boundary instructions defined to isolate untrusted text from the agent's operational prompts.\n
  • Capability inventory: The skill is limited to document generation and formatting (Markdown, JSON, CSV, and export-ready data for QA tools). It possesses no capabilities for arbitrary command execution, network exfiltration, or access to sensitive system files.\n
  • Sanitization: The 'normalization' step described in the workflow handles ambiguity and testability but does not include mechanisms for filtering or sanitizing malicious prompt injections.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 01:42 AM
Security Audit — agent-trust-hub — qa-testcase-from-docs