The Agent Skills Directory

[SAFE]: The skill provides a robust set of tools and instructions for managing AI Gateway configurations, adhering to security best practices throughout its implementation.
[CREDENTIALS_SAFE]: The skill enforces secure credential management. Instructions explicitly prohibit raw API keys in conversations and manifests, requiring the use of tfy-secret:// URI references. The provided tfy-api.sh script handles authentication via environment variables or a local .env file.
[EXTERNAL_DOWNLOADS]: The skill references the official truefoundry Python package and CLI tool. These are recognized as legitimate vendor resources from a well-known service provider.
[COMMAND_EXECUTION]: Shell command usage is limited to the official tfy CLI and local helper scripts designed for API interaction. Destructive operations like resource deletion are intentionally omitted from automated execution, requiring manual user action in the platform dashboard.
[PROMPT_INJECTION]: There are no attempts to override agent behavior or bypass safety guidelines. Instead, the skill actively promotes the configuration of content safety guardrails, including prompt injection detection, at the infrastructure level.

truefoundry-gateway