cabinpanda-automation

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTIONNO_CODE
Full Analysis
  • [EXTERNAL_DOWNLOADS] (LOW): The skill directs users to add an external MCP server endpoint (https://rube.app/mcp) which is not on the trusted repository or organization list.\n- [PROMPT_INJECTION] (LOW): The skill is vulnerable to indirect prompt injection because it instructs the agent to dynamically fetch and follow execution plans and tool schemas from a remote tool.\n
  • Ingestion points: Remote tool output from RUBE_SEARCH_TOOLS as described in SKILL.md.\n
  • Boundary markers: Absent; no instructions are provided to the agent to treat the tool output as untrusted.\n
  • Capability inventory: RUBE_MULTI_EXECUTE_TOOL and RUBE_REMOTE_WORKBENCH which allow execution of remote actions.\n
  • Sanitization: Absent; the instructions explicitly tell the agent to use exact field names and recommended execution plans from search results.\n- [NO_CODE] (SAFE): No executable code files (.py, .js, .sh) are included; the skill consists only of a markdown configuration file.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:32 PM