wallet-profiling
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes structured data from the third-party SolanaTracker API (data.solanatracker.io) to compute and display wallet performance. The ingestion of external, potentially attacker-influenced content like token symbols or addresses without robust isolation creates a potential for indirect prompt injection.\n
- Ingestion points: API response handling in scripts/profile_wallet.py and scripts/compare_wallets.py using the httpx library.\n
- Boundary markers: None. There are no explicit instructions or delimiters used to separate the external data from the system prompt's instructions.\n
- Capability inventory: The skill utilizes network access (httpx.get) to fetch data, but it lacks file-system write or arbitrary code execution capabilities.\n
- Sanitization: The skill performs basic data validation by checking for numeric types and truncating display strings (e.g., token symbols to 8 characters), but it does not fully escape or sanitize strings that could contain malicious instructions.\n- [DATA_EXFILTRATION]: The skill performs network requests to data.solanatracker.io to fetch wallet PnL and transaction data. While this is the primary function of the skill, the use of a non-whitelisted external domain is noted.
Audit Metadata