The Agent Skills Directory

[SAFE]: The skill is purely informational, offering a structured framework (SHOT) for high-quality video production. No malicious patterns or vulnerabilities were found across the provided files.\n- [NO_CODE]: The skill contains no executable code, scripts, or runtime logic. It relies on the AI agent following provided markdown instructions to guide its interactions with external video tools.\n- [PROMPT_INJECTION]: The skill contains explicit safety guardrails that instruct the agent to refuse the creation of non-consensual deepfakes and to ensure AI disclosure compliance, which mitigates risks of misuse.\n- [DATA_EXFILTRATION]: No network exfiltration or sensitive data access was identified. Mentions of API authentication methods (e.g., Bearer tokens) are included for documentation purposes and contain no actual credentials.

runway