ideogram-migration-deep-dive
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEDYNAMIC_CONTEXT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DYNAMIC_CONTEXT_INJECTION]: The skill uses the
!commandsyntax to executenpm list 2>/dev/null | head -10at load time. This is used for environment inspection and is considered benign as it does not access sensitive data or perform unauthorized network operations. - [INDIRECT_PROMPT_INJECTION]: The
validateMigrationfunction in SKILL.md accepts atestPromptsarray that is interpolated into API requests without sanitization, creating a surface for indirect prompt injection if the source data is untrusted. - Ingestion points:
testPromptsparameter in thevalidateMigrationfunction in SKILL.md. - Boundary markers: None; prompts are directly injected into the JSON request body.
- Capability inventory: Bash shell execution, file system modification (Write/Edit), and network access (fetch).
- Sanitization: None; prompt data is processed as raw strings.
Audit Metadata