dropbox-sign-automation

Fail

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • Unverifiable Dependencies (HIGH): The skill requires the addition of a remote MCP server at https://rube.app/mcp. This domain is not within the trusted scope. The security and integrity of any operations depend entirely on this external provider.
  • Indirect Prompt Injection (HIGH): The skill exhibits a high-risk capability tier where external data directly influences side-effect-heavy actions.
  • Ingestion points: Untrusted data enters the agent context via RUBE_SEARCH_TOOLS, which provides "input schemas, recommended execution plans, and known pitfalls."
  • Boundary markers: Absent. No instructions are provided to sanitize or ignore malicious instructions embedded in the tool schemas or plans.
  • Capability inventory: RUBE_MULTI_EXECUTE_TOOL and RUBE_REMOTE_WORKBENCH provide the ability to execute operations on Dropbox Sign and run remote code/tools.
  • Sanitization: Absent. The agent is explicitly told to "Use exact field names and types from the search results."
  • Dynamic Execution (MEDIUM): The skill uses a pattern of dynamic tool discovery and execution. By calling RUBE_GET_TOOL_SCHEMAS and executing TOOL_SLUG_FROM_SEARCH, the agent is performing dynamic loading of executable logic from computed remote paths, which could be manipulated by the remote server to execute unintended commands.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 17, 2026, 08:00 AM