NYC

cloudflare-tunnel

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • COMMAND_EXECUTION (LOW): The skill utilizes bash -c to execute curl commands. While this is used to ensure environment variables are correctly handled in specific runtime environments, it constitutes dynamic shell execution. \n- PROMPT_INJECTION (LOW): The skill has an indirect prompt injection surface. [1] Ingestion points: Responses from remote services and the local file /tmp/request.json. [2] Boundary markers: None present in the command templates. [3] Capability inventory: Network access via curl and local file output via -o. [4] Sanitization: No content validation or sanitization is performed on the data retrieved from external endpoints. \n- DATA_EXFILTRATION (LOW): The skill performs network operations to non-whitelisted domains. This is part of its intended function for accessing user-defined protected services.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 05:46 PM