senpi-smart-money
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external market data which provides a potential surface for indirect prompt injection. \n
- Ingestion points: External trader positioning and leaderboard data are fetched from Hyperliquid via the discovery and leaderboard tools in
scripts/smartmoney.py. \n - Boundary markers: The instructions do not define explicit boundary markers (like XML tags or delimiters) for the data returned by the engine, although the agent is instructed to cite values verbatim. \n
- Capability inventory: The skill uses the
exectool to run a local script engine (scripts/smartmoney.py) that performs network communication with the vendor's MCP server and manages local temporary state. \n - Sanitization: There is no evidence of sanitization or validation of data fields performed on the external content before it is processed by the agent.
Audit Metadata