supabase-automation

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.70). The SKILL.md explicitly instructs the agent to list tables, fetch schemas, and select rows from arbitrary Supabase projects (e.g., SUPABASE_LIST_TABLES, SUPABASE_GET_TABLE_SCHEMAS, SUPABASE_SELECT_FROM_TABLE, and SUPABASE_BETA_RUN_SQL_QUERY), which means it ingests and acts on user-controlled/untrusted database content that can materially influence subsequent queries and actions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly requires connecting at runtime to the MCP endpoint https://rube.app/mcp (used to call RUBE_SEARCH_TOOLS/RUBE_MANAGE_CONNECTIONS) to fetch tool schemas that directly determine agent instructions and is a required dependency.