The Agent Skills Directory

[SAFE]: The skill interacts exclusively with the official Ofox API at api.ofox.ai. This is a vendor-owned domain, and all network activity is restricted to legitimate image generation and model lookup tasks.
[SAFE]: Sensitive credentials, such as the OFOX_API_KEY, are managed securely through the shell environment. The script includes explicit safeguards to prevent the key from being hardcoded, logged, or displayed in output.
[SAFE]: All command-line arguments and external inputs are rigorously validated before processing. JSON payloads are constructed using safe methods like jq --arg, which prevents command injection and payload manipulation.
[SAFE]: File system operations are carefully scoped. The script validates output paths and permissions before initiating billable network requests, and it uses standard, trustworthy tools like curl, jq, and ffmpeg for its operations.
[SAFE]: No malicious patterns such as prompt injection, persistence mechanisms, obfuscation, or unauthorized privilege escalation were detected. The skill maintains a high standard of code quality and transparency.

ofox-image-core