market-research-brief

Pass

Audited by Gen Agent Trust Hub on Nov 22, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied information such as "Topic/domain" and "Data sources" to construct its reports. It does not define boundary markers (such as XML tags or specific delimiters) or provide instructions to the agent to treat this external data as potentially untrusted. This creates a surface where malicious instructions embedded in the analyzed data could influence the agent's behavior.
  • Ingestion points: Found in SKILL.md (Step 1), which identifies user-provided market topics and data sources as primary inputs.
  • Boundary markers: Absent. There are no instructions to wrap or delimit external content.
  • Capability inventory: The skill possesses the capability to write files (PDF, DOCX, PPTX) and execute code via internal libraries for plotting.
  • Sanitization: Not specified; the skill does not mention filtering or escaping characters in the input data.
  • [DYNAMIC_EXECUTION]: The chart_creation_guide.md and SKILL.md (Step 5) specifically mandate that the agent render data visualizations as actual images (PNG) using libraries such as matplotlib or plotly. This workflow requires the agent to generate and execute Python code at runtime to produce the report's visual exhibits. While this is a core feature of a data analysis skill, it represents a dynamic execution pattern.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 22, 2026, 01:11 AM