The Agent Skills Directory

EXTERNAL_DOWNLOADS (LOW): The skill instructs the user to add a remote MCP server endpoint (https://rube.app/mcp). While this is a standard configuration for the Model Context Protocol, it introduces a dependency on a third-party server that is not part of the pre-defined trusted organizations.
REMOTE_CODE_EXECUTION (LOW): The skill utilizes RUBE_MULTI_EXECUTE_TOOL and RUBE_REMOTE_WORKBENCH to perform operations on the Composio/Userlist platform. This involves executing logic on a remote workbench environment.
DATA_EXFILTRATION (LOW): The skill's primary purpose is to process data from 'Userlist', which likely contains PII or sensitive customer data. Because this data is processed via a remote toolkit, there is an inherent risk of data exposure to the tool provider.
PROMPT_INJECTION (LOW): Category 8 (Indirect Prompt Injection). The skill is vulnerable to indirect injection because it dynamically fetches tool schemas and execution plans via RUBE_SEARCH_TOOLS from a remote server.
Ingestion points: Tool schemas and input definitions returned by RUBE_SEARCH_TOOLS (referenced in SKILL.md).
Boundary markers: Absent; the agent is instructed to use the exact field names and types from the search results without specific sanitization instructions.
Capability inventory: RUBE_MULTI_EXECUTE_TOOL and RUBE_REMOTE_WORKBENCH allow the agent to perform actions based on the ingested schemas.
Sanitization: Not present in the skill instructions.

userlist-automation