The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses ssh to execute commands on remote hosts (e.g., ssh <ssh_host> 'curl ...', ssh <relay_host> 'tail ...'). These commands use variables for hostnames and paths derived from configuration files or user input, which can be a vector for command injection if the input sources are untrusted.
[EXTERNAL_DOWNLOADS]: The skill fetches content from external Subscription URLs using curl and from api.ipify.org. While ipify.org is a well-known service, the subscription URLs are user-provided and may contain sensitive tokens.
[PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection because it ingests and processes untrusted data from external sources.
Ingestion points: External content from Subscription URLs (Step 4) and content from remote log files (Step 3) are loaded into the agent's context.
Boundary markers: No explicit delimiters or warnings to ignore embedded instructions within the external data are present.
Capability inventory: The skill possesses the ability to execute shell commands (ssh, curl) and display processed data back to the user.
Sanitization: There is no evidence of sanitization or validation of the content retrieved from external URLs or log files before processing.

ray-nodecheck