The Agent Skills Directory

[DATA_EXFILTRATION]: The skill transmits user-provided code to https://ray.tinte.dev/api/v1/screenshot to generate images. This presents a potential risk if the code being processed contains sensitive information, such as API keys or private credentials, which are then exposed to the external service.
[COMMAND_EXECUTION]: The skill instructions specify the use of the curl command to interact with the remote API and the open command to display the generated PNG file locally.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through its ingestion of untrusted data. \n
Ingestion points: The skill ingests arbitrary code provided by the user or read from files via the code parameter. \n
Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat the code as data only and ignore any embedded instructions. \n
Capability inventory: The skill possesses network communication capabilities (curl) and local command execution capabilities (open). \n
Sanitization: No sanitization or validation of the input code is performed before it is sent to the API or processed by the agent.

ray