quicknode-core-workflow-a
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external, untrusted sources by listening to blockchain events and querying transaction receipts.
- Ingestion points: Untrusted data enters the agent context via
ethers.WebSocketProviderevent listeners andethers.JsonRpcProviderqueries as shown inSKILL.md. - Boundary markers: There are no explicit boundary markers or instructions to the agent to disregard potential instructions embedded in transaction metadata or event logs.
- Capability inventory: The skill is granted significant capabilities, including the ability to write/edit files and execute shell commands via
Bash(npm:*)andBash(curl:*). - Sanitization: The example code snippets demonstrate logging raw event data (e.g.,
from,to,value) to the console without explicit validation or sanitization.
Audit Metadata