The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs users to add a remote MCP server at https://rube.app/mcp. This server provides the core logic and schemas for all Segment automation tools used by the skill. Using a non-trusted, third-party endpoint for tool definitions allows the remote service to define the agent's behavior and potentially capture or exfiltrate sensitive data processed during Segment operations.
[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes data from external sources that could contain malicious instructions designed to manipulate the agent's behavior.
Ingestion points: Potentially untrusted data enters the agent's context through Segment event properties, user traits, and source schema settings retrieved via SEGMENT_LIST_SCHEMA_SETTINGS_IN_SOURCE.
Boundary markers: The instructions do not specify any delimiters or safety warnings to ensure the agent ignores instructions embedded within the processed Segment data.
Capability inventory: The skill possesses powerful capabilities, including modifying source configurations (SEGMENT_UPDATE_SOURCE), merging user identities (SEGMENT_ALIAS), and performing batch data operations (SEGMENT_BATCH).
Sanitization: There is no evidence of input validation, escaping, or sanitization of data before it is interpolated into Segment tool calls or processed by the agent.

segment-automation