The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill installs the nansen-cli package from the NPM registry. The package is maintained by the well-known organization Nansen AI on GitHub.
[DATA_EXFILTRATION]: The skill accesses ~/.nansen/.env to retrieve the NANSEN_WALLET_PASSWORD. This constitutes access to a sensitive path, but it is performed as part of the tool's standard credential management for executing trades.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it interpolates user-supplied data into shell commands without sanitization.
Ingestion points: User-provided token symbols, contract addresses, and amounts in SKILL.md; output from the nansen trade quote command.
Boundary markers: No delimiters are present to isolate untrusted data.
Capability inventory: The skill allows executing shell commands through the nansen binary via the Bash(nansen:*) tool.
Sanitization: No validation or escaping is applied to user-controlled parameters before command execution.
[COMMAND_EXECUTION]: The skill executes shell commands, specifically sourcing an environment file and running the nansen trading CLI, which is the primary purpose of the skill.

nansen-trading