segment-automation

Pass

Audited by Gen Agent Trust Hub on Feb 20, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • External Downloads (LOW): The skill directs users to connect to a third-party MCP server at https://rube.app/mcp.
  • Evidence: Found in the 'Setup' section of SKILL.md.
  • Context: While this is a functional requirement for the skill to operate, the domain is not on the pre-approved trusted list.
  • Prompt Injection (LOW): The skill is susceptible to indirect prompt injection due to its handling of untrusted external data during Segment operations.
  • Ingestion points: Data enters the agent context through traits and properties parameters in tools like SEGMENT_TRACK, SEGMENT_IDENTIFY, SEGMENT_GROUP, and SEGMENT_PAGE.
  • Boundary markers: There are no delimiters or explicit instructions to ignore embedded commands within the data objects being sent to Segment.
  • Capability inventory: The skill possesses significant write capabilities, including event tracking, user identification, and source configuration management (SEGMENT_UPDATE_SOURCE).
  • Sanitization: The skill lacks any description of sanitization or validation for the data interpolated into Segment API calls.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 20, 2026, 08:18 AM