nansen-wallet-deep-dive
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Installs the
nansen-clipackage from the Node Package Manager (npm) to provide the core functionality of the skill. - [COMMAND_EXECUTION]: Executes a series of subcommands under
nansen research profilerto extract labels, balances, PnL summaries, transaction history, perpetual positions, and counterparty data for specified blockchain addresses. - [INDIRECT_PROMPT_INJECTION]: The skill processes external blockchain data and user-provided inputs (wallet addresses and chain names) which constitutes an injection surface.
- Ingestion points: Variable interpolation of
$ADDRand$CHAINwithin shell commands inSKILL.md. - Boundary markers: None present to delineate external data from agent instructions.
- Capability inventory: Execution is limited to the
nansenbinary viaBash(nansen:*)as defined in theallowed-toolsmetadata. - Sanitization: Not explicitly defined; the skill relies on the
nansen-clito handle inputs correctly.
Audit Metadata