The Agent Skills Directory

[DATA_EXPOSURE]: The skill identifies and accesses potentially sensitive files such as ~/.nansen/.env and .credentials to migrate stored passwords. This is the primary intended purpose of the skill and is performed by the vendor's own CLI tool. The instructions explicitly command the agent to suppress all output using > /dev/null 2>&1 and unset variables immediately after use to ensure secrets are not captured in conversation history or logs.
[COMMAND_EXECUTION]: The skill uses potentially dangerous commands like nansen wallet export --reveal. However, this is used strictly for verification and is coupled with mandatory shell redirection to discard all output, ensuring that private keys are used only for exit-code validation and are never visible to the agent or user.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external files (~/.nansen/.env). It mitigates risks by sourcing the file directly into the shell environment for the migration command rather than reading the file content into the agent's natural language context, preventing the agent from being influenced by any malicious instructions that might be embedded in the file.

nansen-wallet-keychain-migration