clade-load-scale
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the processing of external, potentially untrusted data through the Claude API via message batches and request queues.\n
- Ingestion points: Data inputs
itemsandinputsinSKILL.mdare passed directly to API calls.\n - Boundary markers: None present in the code snippets.\n
- Capability inventory: The skill demonstrates usage of the Anthropic SDK for message creation and batch processing.\n
- Sanitization: No sanitization or input filtering is demonstrated in the examples.
Audit Metadata