polymarket-scanner
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill performs network requests to external domains to fetch market metadata and real-time pricing information.
- Evidence:
scan_markets.pyandget_prices.pymake GET requests togamma-api.polymarket.comandclob.polymarket.com. These domains belong to the official Polymarket service. - [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted, user-generated questions and descriptions from the Polymarket API which could contain malicious instructions.
- Ingestion points:
scan_markets.pyretrieves market text from the external Gamma API. - Boundary markers:
SKILL.mdincludes an explicit caution warning the agent that market data is untrusted and should not be interpreted as instructions. - Capability inventory: The skill has network read access and executes specific Python scripts. It lacks the ability to write to the file system or execute arbitrary shell commands.
- Sanitization:
scan_markets.pyuses asanitize_textfunction to strip control characters and truncate long strings, reducing the risk of certain injection vectors.
Audit Metadata