moonpay-scout
Pass
Audited by Gen Agent Trust Hub on Mar 21, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection. It ingest market titles and descriptions from Polymarket and Kalshi to perform 'Alpha Analysis' and 'Arb Math'. An attacker could create a market with instructions hidden in the question or description to manipulate the agent into executing specific trades or bypassing its conviction checks.
- Ingestion points: The
{{args}}parameter and external market data retrieved via the MoonPay CLI. - Boundary markers: None identified in the prompt templates.
- Capability inventory: The skill can execute financial trades via the
mp prediction-market position buycommand. - Sanitization: No evidence of validation or filtering for the external market content before processing.
- [COMMAND_EXECUTION]: The skill uses the MoonPay CLI to execute trades on the Polygon network. While this is the intended functionality, the parameters for these commands (token IDs, prices, and sizes) are dynamically generated based on potentially untrusted external data.
Audit Metadata