The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to interact with the shoplazza CLI tool to perform store management operations. It provides a mapping of user intents to specific CLI commands such as webhook create, webhook list, and webhook delete.
[INDIRECT_PROMPT_INJECTION]: The skill defines an attack surface where user-supplied data, specifically webhook URLs and event topics, are interpolated into CLI commands.
Ingestion points: User-provided inputs for the --data and --params flags in the SKILL.md file.
Boundary markers: The skill explicitly instructs the agent to use AskUserQuestion if required fields are missing and to never fabricate data. It also references a common safety protocol for confirmation.
Capability inventory: Execution of shell-based CLI commands via the shoplazza binary.
Sanitization: The skill relies on the shoplazza CLI's own schema validation and the safety protocols defined in the referenced common skill to handle untrusted input.

shoplazza-webhook