supabase-skill

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill includes explicit CLI examples that pass SUPABASE_ACCESS_TOKEN inline (e.g., --env SUPABASE_ACCESS_TOKEN=your_token), which would lead an agent to embed real secret values verbatim in generated commands (exfiltration risk), even though some examples show safe env var usage.

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

• Potentially malicious external URL detected (high risk: 0.80). The SKILL contains a runtime installation command that fetches and executes remote code—"npx -y @supabase/mcp-server-supabase@latest"—which would install/run a remote MCP server package (executing code that can control agent prompts/context), so this is a risky runtime dependency.