prompts
Pass
Audited by Gen Agent Trust Hub on Apr 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a standard management tool for TrueFoundry services and does not contain any malicious behavior or security bypasses.
- [COMMAND_EXECUTION]: Provides authenticated helper scripts (
tfy-api.sh,tfy-version.sh) for interacting with the TrueFoundry API. These scripts are properly scoped and include validation to prevent common attacks like path traversal. - [EXTERNAL_DOWNLOADS]: References the
truefoundryPython package and various official container images from trusted registries (AWS ECR, GHCR, NVIDIA). These are documented for legitimate service deployment and management. - [CREDENTIALS_UNSAFE]: Promotes secure secret management by using environment variables,
.envfiles, andtfy-secret://references instead of hardcoding sensitive data. - [PROMPT_INJECTION]: While the skill manages LLM prompts (an indirect injection surface), it explicitly includes warnings to review prompt content from untrusted sources before creation.
Audit Metadata