adobe-rate-limits
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides patterns for batch processing of user-provided data, creating a surface for indirect prompt injection attacks where the agent might interpret data as instructions.
- Ingestion points: The
batchFireflyGeneratefunction inSKILL.mdaccepts apromptsarray, which serves as an entry point for untrusted external strings. - Boundary markers: The code examples provided do not include delimiters or specific instructions for the agent to ignore or isolate commands that may be embedded within the user-provided prompts.
- Capability inventory: The skill's frontmatter configuration allows the use of
Read,Write, andEdittools, which could be leveraged by a successful injection attack to modify files or access data. - Sanitization: The implementation logic lacks input validation, filtering, or escaping mechanisms to prevent the agent from inadvertently executing instructions found within the processed strings.
Audit Metadata