The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses a bundled script ${CLAUDE_PLUGIN_ROOT}/scripts/gb-call to interact with GrowthBook's API. This is the primary mechanism for its stated purpose.
[CREDENTIALS_UNSAFE]: References the use of GB_API_KEY, which is expected to be provided via environment variables or a configuration file (~/.config/growthbook/.env). The skill follows standard security practices by using environment-based secrets rather than hardcoding credentials.
[DATA_EXFILTRATION]: All network requests are directed to GrowthBook's official API domains. There is no evidence of data being sent to unauthorized third-party services.
[EXTERNAL_DOWNLOADS]: Uses system commands like open or xdg-open to navigate the user to the GrowthBook web interface for manual review or conflict resolution. These actions target well-known vendor domains.
[PROMPT_INJECTION]: Instructions are focused on operational logic and do not contain attempts to bypass agent safety filters or override system instructions.

flag-publish