typescript-any-eliminator
Pass
Audited by Gen Agent Trust Hub on Nov 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill refactors untrusted TypeScript code and executes local validation commands, which presents a surface for indirect prompt injection.
- Ingestion points: Untrusted TypeScript source files are read as part of the primary workflow in
SKILL.md. - Boundary markers: There are no instructions to use markers or safety prompts to isolate code content from the agent's instruction stream.
- Capability inventory: The skill modifies source files and executes shell commands, such as
tscor project-specific type-check wrappers, to verify changes. - Sanitization: No input validation is performed on the source code to detect or neutralize embedded malicious instructions.
Audit Metadata