The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a terminal command autonew $PROVIDER where the $PROVIDER variable is supplied as input. Direct interpolation of variables into shell commands without explicit sanitization or escaping presents a potential command injection surface.
[PROMPT_INJECTION]: The skill exposes a surface for indirect prompt injection by accepting an untrusted variable ($PROVIDER) used in an execution context. Ingestion points: The $PROVIDER variable defined in the Execution section of SKILL.md. Boundary markers: None present; input is used directly in the shell string. Capability inventory: The skill uses the autonew command-line utility to interact with provider sessions. Sanitization: No input validation or sanitization logic is defined within the skill file.

autonew