The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell tools like curl and jq to manage project tasks. It employs jq to safely build JSON payloads, which is a best practice to prevent command injection or malformed data issues when handling variables like task titles and descriptions.
[EXTERNAL_DOWNLOADS]: The skill communicates with an API endpoint defined by the $BASE_URL variable. This network activity is restricted to the skill's primary function of task orchestration and project management.
[DATA_EXFILTRATION]: There is no evidence of sensitive data being sent to unauthorized external domains. The skill uses project-specific credentials (AUTH_HEADER) to interact with its intended backend service.
[PROMPT_INJECTION]: The skill provides clear, task-oriented instructions without any attempts to bypass safety filters, extract system prompts, or override the agent's core behavioral guidelines.
[DATA_EXPOSURE]: The skill reads from a shared configuration file (../squad/shared.md) to resolve environment variables and authentication tokens. This is a standard and safe method for managing configuration in this agent environment.

squad-kickstart