typesafe-ai
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists entirely of markdown documentation and reference materials for the TypeSafe AI decision model.
- [SAFE]: No executable code, shell scripts, or malicious binaries are included in the skill package.
- [CREDENTIALS_UNSAFE]: Secret management follows industry best practices by instructing the use of environment variables (TYPESAFE_API_KEY) and providing placeholders in examples rather than hardcoding sensitive data.
- [INDIRECT_PROMPT_INJECTION]: The documentation explicitly addresses the risk of adversarial text within the processed state data and provides specific 'guardrails' and 'citation_check' cookbooks to help developers detect and mitigate such inputs using the model's own judgment capabilities.
Audit Metadata