The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using curl to interact with the Karma API endpoint at gapapi.karmahq.xyz. These commands are used to initialize authentication and verify codes. It also uses python3 -m json.tool for formatting API responses.
[PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection because it processes untrusted user data (email addresses and verification codes) which are then interpolated into shell commands. * Ingestion points: User-provided email and verification code (Step 2 and Step 3 of SKILL.md). * Boundary markers: No explicit delimiters or instructions are provided to the agent to treat the user input as data rather than command components. * Capability inventory: The skill uses curl for network requests and python3 for processing output. * Sanitization: There is no mention of sanitizing or escaping the user-provided strings before they are used in the command-line interface.

setup-agent