business-skills

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill provides legitimate analytical tools for business and innovation research. All analyzed scripts use standard Python libraries for mathematical modeling and data processing without performing unauthorized network or system operations.- [COMMAND_EXECUTION]: In market-analysis-guide/SKILL.md, the skill uses matplotlib.pyplot.savefig to generate and save a horizontal bar chart to market_share.png. This is a routine operation for a data visualization utility and is not considered a security risk.- [INDIRECT_PROMPT_INJECTION]: The skill exposes surfaces for processing external data through several analytical functions.
  • Ingestion points: Resource lists and firm performance metrics are passed as arguments to functions like vrio_analysis and compute_innovation_scorecard across various sub-skills.
  • Boundary markers: None explicitly defined in the logic for the agent to separate untrusted data.
  • Capability inventory: Mathematical calculations, data frame manipulation, and image file creation (PNG).
  • Sanitization: None. The attack surface is limited as the data is used for structured computation rather than influence over the agent's core instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 09:31 AM