typesafe-ai
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses a command-line interface tool named
typesafeto process and score local documents. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-provided text and Markdown files through the
typesafeCLI and TypeSafe API, which could potentially contain malicious instructions intended to influence the behavior of the agent or the underlying model. - Ingestion points:
typesafe [--rubric FILE] DOCUMENT...(SKILL.md) - Boundary markers: None explicitly defined in the skill instructions to delimit untrusted document content.
- Capability inventory: Local command execution (
typesafeCLI), local file reading, and network transmission of document data to the TypeSafe API. - Sanitization: No specific sanitization or filtering logic is described for the input documents.
Audit Metadata