adversarial-ts-patterns
Pass
Audited by Gen Agent Trust Hub on Jul 20, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates as a static code analysis tool that utilizes AI subagents to evaluate target code against a predefined set of architectural guidelines. There is no evidence of network exfiltration, hardcoded credentials, or unauthorized system modifications. Documentation links reference trusted domains such as react.dev and typescriptlang.org.
- [PROMPT_INJECTION]: The skill's architecture involves ingesting external code for review, which introduces a potential surface for indirect prompt injection. This is a functional risk inherent to the tool's purpose as a code auditor.
- Ingestion points: The target code is ingested into the
references/reviewer-prompt.mdtemplate through the{{TARGET_CONTENT_OR_PATHS}}variable. - Boundary markers: The prompt template lacks explicit delimiters or specific warnings to distinguish input code from analysis instructions.
- Capability inventory: The analysis is performed by a Task subagent with standard platform constraints and no specialized dangerous capabilities.
- Sanitization: No content filtering or sanitization is applied to the analyzed code prior to its inclusion in the subagent's prompt.
Audit Metadata