The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill uses npx to fetch and run the starkfi package from the public NPM registry at runtime.
[COMMAND_EXECUTION]: It utilizes the Bash tool to execute shell commands. The use of a wildcard * in the allowed tool definition npx starkfi@latest send * indicates that user input is passed directly to the shell.
[REMOTE_CODE_EXECUTION]: The execution pattern npx starkfi@latest downloads and runs remote code, which constitutes a remote code execution vector if the package source is compromised or if the package itself contains malicious logic.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it interpolates user-provided data such as recipient and token directly into shell commands.
Ingestion points: User input for amount, token, and recipient in SKILL.md.
Boundary markers: Absent; there are no delimiters or instructions to ignore embedded commands in the input.
Capability inventory: Uses Bash for shell command execution.
Sanitization: Absent; the skill does not specify any validation or escaping for the input strings before they are executed in the shell.

send