The Agent Skills Directory

[COMMAND_EXECUTION]: The skill defines a shell function api() that calls a local script (api-request.sh) to interact with Spotify Ads API services. This is the primary mechanism for the skill's functionality and is used for resource discovery and configuration.
[INDIRECT_PROMPT_INJECTION]: The skill ingests user-supplied identifiers (such as business_id and ad_account_id) and implementation details to construct shell-based API calls. This creates a potential surface for indirect injection; however, the skill explicitly mitigates this risk by instructing the agent to confirm all IDs before mutation and to present exact plans before execution.
[DATA_EXPOSURE]: The instructions include specific safeguards for handling sensitive data, requiring the agent to redact CAPI tokens and hashed user identifiers (email, phone) from all chat outputs and logs. It also advises using external secret managers for token storage rather than exposing them in the conversation.

measurement-setup