the-hypothesis-engine

Pass

Audited by Gen Agent Trust Hub on Aug 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is entirely instructional and serves as a template generator for business experiments. It contains no executable code, shell scripts, or remote network operations.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests content from local files such as .agents/product-context.md and references/bayesian-testing.md. While these represent external data ingestion points, the skill's primary function is text generation within a structured brief, and it does not possess any dangerous capabilities (e.g., file writing, command execution, or network exfiltration) that could be exploited via malicious content in those files.
  • [DATA_EXPOSURE]: The instructions require reading business-related information from .agents/product-context.md. This is a standard pattern for context-aware agents to personalize their output and does not involve exfiltrating this data to external servers.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 14, 2026, 12:03 PM