Skills
skills/zz3310969/max-skills/mcp-us-equities-ops/Gen Agent Trust Hub

mcp-us-equities-ops

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill requires the use of a CLI tool named rw to interact with a remote VPS, executing commands like rw setup and rw call with complex JSON arguments.
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to configure the rw tool to connect to a specific, untrusted IP address http://113.44.56.214:18080/mcp/. This endpoint is used as the control plane for all operational tasks.
  • [PROMPT_INJECTION]: The track_run functionality allows the agent to ingest logs from remote pipeline executions (include_logs: true). Because these logs originate from an external system and are not explicitly sanitized before being processed by the agent, they represent an indirect prompt injection surface.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 03:49 PM