The Agent Skills Directory

[PROMPT_INJECTION]: The skill instructions provide clear, functional guidance for media tasks without attempting to bypass agent safety filters or override system behavior.
[CREDENTIALS_UNSAFE]: Secure API key management is enforced by directing the agent to use environment variables or project-specific .env files, with explicit instructions for the agent to avoid direct handling of secrets.
[EXTERNAL_DOWNLOADS]: Dependencies are limited to the official videodb Python SDK and the standard python-dotenv package, which are installed from trusted registries.
[COMMAND_EXECUTION]: Use of the system shell is restricted to legitimate lifecycle management of the included WebSocket listener script and standard Python interactions with the official SDK.
[DATA_EXFILTRATION]: Outbound network activity is directed exclusively to authorized VideoDB service endpoints required for the skill's primary functionality.
[SAFE]: Local file access and storage are consistent with the skill's purpose, limited to project configuration and temporary event logging for capture sessions.

videodb