israeli-market-fit

Pass

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user data (product descriptions and pricing) to generate a market-fit report.
  • Ingestion points: User-provided product details and price points processed in SKILL.md and SKILL_HE.md.
  • Boundary markers: The skill relies on structured sections and bundled reference files (references/segments.md) to scope the analysis, though it lacks explicit delimiters for user input.
  • Capability inventory: The agent is instructed to execute a local Python script (scripts/market-sizing.py) to calculate market metrics.
  • Sanitization: No explicit sanitization or filtering is performed on user-supplied product descriptions before processing.
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute a local Python script (scripts/market-sizing.py) via the command line to perform market sizing calculations. The script uses only standard Python libraries (argparse, sys) and does not perform network operations or access sensitive files.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 21, 2026, 02:22 PM
Security Audit — agent-trust-hub — israeli-market-fit