The Agent Skills Directory

[SAFE]: The skill interacts exclusively with the vendor's official API at anygen.io. All network requests are performed using the requests library with security measures such as disabling redirects and using structured authentication headers.
[SAFE]: Credential management follows security best practices by allowing API keys to be provided via environment variables or a configuration file located at ~/.config/anygen/config.json with restricted filesystem permissions (chmod 600) to protect the secret.
[SAFE]: File handling is strictly scoped to user-provided reference files and the designated workspace directory. The instructions mandate explicit user consent before any file is read or uploaded to the external service, preventing unauthorized data exposure.
[SAFE]: The use of background monitoring via sessions_spawn is restricted to status polling and notifying the user of task completion. It uses internal task identifiers generated by the service, following the platform's intended autonomy and monitoring model.
[SAFE]: No malicious patterns such as prompt injection, obfuscation, or unauthorized data exfiltration were detected within the instructions or the accompanying Python scripts. The skill includes instructions to summarize data rather than relaying raw sensitive content, providing a measure of protection against indirect prompt injection.

anygen-website