Supabase RLS policy is correct but agent queries time out with 30s latency

A Supabase query behind a row-level security policy takes 30+ seconds for a signed-in user. Without RLS the same query runs in 40ms. EXPLAIN shows the policy's USING clause forces a sequential scan over 2M rows per call.

context

Policy: `USING (auth.uid() = user_id OR auth.uid() IN (SELECT admin_id FROM team_admins WHERE team_id = X.team_id))`. Table has 2M rows, user_id is indexed, team_admins has 50k rows.

goal

Rewrite the RLS policy to be index-compatible. Explain how Postgres's query planner handles nested-subquery policies and what change unblocks index use.

constraints

Must preserve exact authorization semantics.