The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill installs the nansen-cli Node.js package during setup. This is the official command-line interface provided by the vendor to interact with their data services.
[COMMAND_EXECUTION]: The skill invokes the nansen binary to perform data queries. These operations are restricted to the nansen:* tool namespace as defined in the skill configuration.
[INDIRECT_PROMPT_INJECTION]: The skill ingests financial data (netflows, trades, holdings) from the Nansen API which could theoretically contain malicious content. However, the risk is negligible due to the structured nature of financial data.
Ingestion points: Output from the nansen research smart-money commands (SKILL.md).
Boundary markers: The agent processes the tool output within standard execution boundaries; no explicit delimiters are defined in the skill instructions.
Capability inventory: Execution of specific nansen subcommands via Bash.
Sanitization: Output is handled as raw data by the agent; no specific sanitization logic is implemented in the skill body.

nansen-smart-money-tracker