mike/backend/schema.sql

-- Mike Supabase schema
-- Use this for a fresh Supabase database. Existing deployments should instead
-- apply the dated incremental migration files in backend/migrations that are
-- newer than the version of Mike they currently have deployed.

create extension if not exists "pgcrypto";

-- ---------------------------------------------------------------------------
-- User profiles
-- ---------------------------------------------------------------------------

create table if not exists public.user_profiles (
  id uuid primary key default gen_random_uuid(),
  user_id uuid not null unique references auth.users(id) on delete cascade,
  display_name text,
  organisation text,
  tier text not null default 'Free',
  message_credits_used integer not null default 0,
  credits_reset_date timestamptz not null default (now() + interval '30 days'),
  title_model text,
  tabular_model text not null default 'gemini-3-flash-preview',
  quote_model text,
  mfa_on_login boolean not null default false,
  legal_research_us boolean not null default true,
  created_at timestamptz not null default now(),
  updated_at timestamptz not null default now()
);

create index if not exists idx_user_profiles_user
  on public.user_profiles(user_id);

create or replace function public.handle_new_user()
returns trigger
language plpgsql
security definer
set search_path = public
as $$
begin
  insert into public.user_profiles (user_id)
  values (new.id)
  on conflict (user_id) do nothing;
  return new;
exception when others then
  -- Never block signup if the profile insert fails.
  return new;
end;
$$;

drop trigger if exists on_auth_user_created on auth.users;
create trigger on_auth_user_created
  after insert on auth.users
  for each row execute procedure public.handle_new_user();

create table if not exists public.user_api_keys (
  id uuid primary key default gen_random_uuid(),
  user_id uuid not null references auth.users(id) on delete cascade,
  provider text not null check (provider in ('claude', 'gemini', 'openai', 'openrouter', 'courtlistener')),
  encrypted_key text not null,
  iv text not null,
  auth_tag text not null,
  created_at timestamptz not null default now(),
  updated_at timestamptz not null default now(),
  unique(user_id, provider)
);

create index if not exists idx_user_api_keys_user
  on public.user_api_keys(user_id);

alter table public.user_api_keys enable row level security;

create table if not exists public.user_mcp_connectors (
  id uuid primary key default gen_random_uuid(),
  user_id uuid not null references auth.users(id) on delete cascade,
  name text not null,
  transport text not null default 'streamable_http'
    check (transport in ('streamable_http')),
  server_url text not null,
  auth_type text not null default 'none'
    check (auth_type in ('none', 'bearer', 'oauth')),
  enabled boolean not null default true,
  tool_policy jsonb not null default '{}'::jsonb,
  encrypted_auth_config text,
  auth_config_iv text,
  auth_config_tag text,
  created_at timestamptz not null default now(),
  updated_at timestamptz not null default now()
);

create index if not exists idx_user_mcp_connectors_user
  on public.user_mcp_connectors(user_id);

alter table public.user_mcp_connectors enable row level security;

create table if not exists public.user_mcp_oauth_tokens (
  id uuid primary key default gen_random_uuid(),
  connector_id uuid not null references public.user_mcp_connectors(id) on delete cascade,
  encrypted_access_token text,
  access_token_iv text,
  access_token_tag text,
  encrypted_refresh_token text,
  refresh_token_iv text,
  refresh_token_tag text,
  token_type text,
  scope text,
  expires_at timestamptz,
  authorization_server text,
  token_endpoint text,
  client_id text,
  encrypted_client_secret text,
  client_secret_iv text,
  client_secret_tag text,
  resource text,
  created_at timestamptz not null default now(),
  updated_at timestamptz not null default now(),
  unique(connector_id)
);

alter table public.user_mcp_oauth_tokens enable row level security;

create table if not exists public.user_mcp_oauth_states (
  id uuid primary key default gen_random_uuid(),
  user_id uuid not null references auth.users(id) on delete cascade,
  connector_id uuid not null references public.user_mcp_connectors(id) on delete cascade,
  state_hash text not null unique,
  encrypted_state_config text not null,
  state_config_iv text not null,
  state_config_tag text not null,
  expires_at timestamptz not null,
  created_at timestamptz not null default now()
);

create index if not exists idx_user_mcp_oauth_states_expires
  on public.user_mcp_oauth_states(expires_at);

alter table public.user_mcp_oauth_states enable row level security;

create table if not exists public.user_mcp_connector_tools (
  id uuid primary key default gen_random_uuid(),
  connector_id uuid not null references public.user_mcp_connectors(id) on delete cascade,
  tool_name text not null,
  openai_tool_name text not null,
  title text,
  description text,
  input_schema jsonb not null default '{"type":"object","properties":{}}'::jsonb,
  output_schema jsonb,
  annotations jsonb not null default '{}'::jsonb,
  enabled boolean not null default true,
  requires_confirmation boolean not null default false,
  last_seen_at timestamptz not null default now(),
  created_at timestamptz not null default now(),
  updated_at timestamptz not null default now(),
  unique(connector_id, tool_name),
  unique(openai_tool_name)
);

create index if not exists idx_user_mcp_connector_tools_connector
  on public.user_mcp_connector_tools(connector_id);

alter table public.user_mcp_connector_tools enable row level security;

create table if not exists public.user_mcp_tool_audit_logs (
  id uuid primary key default gen_random_uuid(),
  user_id uuid not null references auth.users(id) on delete cascade,
  connector_id uuid not null references public.user_mcp_connectors(id) on delete cascade,
  tool_id uuid references public.user_mcp_connector_tools(id) on delete set null,
  tool_name text not null,
  openai_tool_name text not null,
  status text not null check (status in ('ok', 'error')),
  error_message text,
  duration_ms integer not null default 0,
  result_size_chars integer not null default 0,
  created_at timestamptz not null default now()
);

create index if not exists idx_user_mcp_tool_audit_logs_user_created
  on public.user_mcp_tool_audit_logs(user_id, created_at desc);

alter table public.user_mcp_tool_audit_logs enable row level security;

-- ---------------------------------------------------------------------------
-- Projects and documents
-- ---------------------------------------------------------------------------

create table if not exists public.projects (
  id uuid primary key default gen_random_uuid(),
  user_id text not null,
  name text not null,
  cm_number text,
  visibility text not null default 'private',
  shared_with jsonb not null default '[]'::jsonb,
  created_at timestamptz not null default now(),
  updated_at timestamptz not null default now()
);

create index if not exists idx_projects_user
  on public.projects(user_id);

create index if not exists projects_shared_with_idx
  on public.projects using gin (shared_with);

create table if not exists public.project_subfolders (
  id uuid primary key default gen_random_uuid(),
  project_id uuid not null references public.projects(id) on delete cascade,
  user_id text not null,
  name text not null,
  parent_folder_id uuid references public.project_subfolders(id) on delete cascade,
  created_at timestamptz not null default now(),
  updated_at timestamptz not null default now()
);

create index if not exists idx_project_subfolders_project
  on public.project_subfolders(project_id);

create table if not exists public.documents (
  id uuid primary key default gen_random_uuid(),
  project_id uuid references public.projects(id) on delete cascade,
  user_id text not null,
  status text not null default 'pending',
  folder_id uuid references public.project_subfolders(id) on delete set null,
  created_at timestamptz not null default now(),
  updated_at timestamptz not null default now()
);

create index if not exists idx_documents_user_project
  on public.documents(user_id, project_id);

create index if not exists idx_documents_project_folder
  on public.documents(project_id, folder_id);

create table if not exists public.document_versions (
  id uuid primary key default gen_random_uuid(),
  document_id uuid not null references public.documents(id) on delete cascade,
  storage_path text,
  pdf_storage_path text,
  source text not null default 'upload',
  version_number integer,
  filename text,
  file_type text,
  size_bytes integer,
  page_count integer,
  deleted_at timestamptz,
  deleted_by uuid,
  created_at timestamptz not null default now(),
  constraint document_versions_source_check
    check (source = any (array[
      'upload'::text,
      'user_upload'::text,
      'assistant_edit'::text,
      'user_accept'::text,
      'user_reject'::text,
      'generated'::text
    ]))
);

create index if not exists document_versions_document_id_idx
  on public.document_versions(document_id, created_at desc);

create index if not exists document_versions_active_document_id_idx
  on public.document_versions(document_id, created_at desc)
  where deleted_at is null;

create index if not exists document_versions_doc_vnum_idx
  on public.document_versions(document_id, version_number);

do $$
begin
  if not exists (
    select 1
    from pg_constraint
    where conname = 'document_versions_doc_version_unique'
      and conrelid = 'public.document_versions'::regclass
  ) then
    alter table public.document_versions
      add constraint document_versions_doc_version_unique
      unique (document_id, version_number);
  end if;
end;
$$;

alter table public.documents
  add column if not exists current_version_id uuid
  references public.document_versions(id) on delete set null;

create table if not exists public.document_edits (
  id uuid primary key default gen_random_uuid(),
  document_id uuid not null references public.documents(id) on delete cascade,
  chat_message_id uuid,
  version_id uuid not null references public.document_versions(id) on delete cascade,
  change_id text not null,
  del_w_id text,
  ins_w_id text,
  deleted_text text not null default '',
  inserted_text text not null default '',
  context_before text,
  context_after text,
  status text not null default 'pending'
    check (status = any (array[
      'pending'::text,
      'accepted'::text,
      'rejected'::text
    ])),
  created_at timestamptz not null default now(),
  resolved_at timestamptz
);

create index if not exists document_edits_document_id_idx
  on public.document_edits(document_id, created_at desc);

create index if not exists document_edits_message_id_idx
  on public.document_edits(chat_message_id);

create index if not exists document_edits_version_id_idx
  on public.document_edits(version_id);

-- ---------------------------------------------------------------------------
-- Workflows
-- ---------------------------------------------------------------------------

create table if not exists public.workflows (
  id uuid primary key default gen_random_uuid(),
  user_id text,
  title text not null,
  type text not null,
  prompt_md text,
  columns_config jsonb,
  practice text,
  is_system boolean not null default false,
  created_at timestamptz not null default now()
);

create index if not exists idx_workflows_user
  on public.workflows(user_id);

create table if not exists public.hidden_workflows (
  id uuid primary key default gen_random_uuid(),
  user_id text not null,
  workflow_id text not null,
  created_at timestamptz not null default now(),
  unique(user_id, workflow_id)
);

create index if not exists idx_hidden_workflows_user
  on public.hidden_workflows(user_id);

create table if not exists public.workflow_shares (
  id uuid primary key default gen_random_uuid(),
  workflow_id uuid not null references public.workflows(id) on delete cascade,
  shared_by_user_id text not null,
  shared_with_email text not null,
  allow_edit boolean not null default false,
  created_at timestamptz not null default now(),
  constraint workflow_shares_workflow_email_unique
    unique(workflow_id, shared_with_email)
);

create index if not exists workflow_shares_workflow_id_idx
  on public.workflow_shares(workflow_id);

create index if not exists workflow_shares_email_idx
  on public.workflow_shares(shared_with_email);

create or replace function public.get_workflows_overview(
  p_user_id text,
  p_user_email text default null,
  p_type text default null
)
returns table (
  id uuid,
  user_id text,
  title text,
  type text,
  prompt_md text,
  columns_config jsonb,
  practice text,
  is_system boolean,
  created_at timestamptz,
  allow_edit boolean,
  is_owner boolean,
  shared_by_name text
)
language sql
stable
as $$
  with owned as (
    select
      w.*,
      true as allow_edit,
      true as is_owner,
      null::text as shared_by_name,
      0 as sort_bucket
    from public.workflows w
    where w.user_id = p_user_id
      and w.is_system = false
      and (p_type is null or w.type = p_type)
  ),
  shared as (
    select
      w.*,
      ws.allow_edit,
      false as is_owner,
      nullif(trim(up.display_name), '') as shared_by_name,
      1 as sort_bucket
    from public.workflow_shares ws
    join public.workflows w
      on w.id = ws.workflow_id
    left join public.user_profiles up
      on up.user_id::text = ws.shared_by_user_id
    where lower(ws.shared_with_email) = lower(coalesce(p_user_email, ''))
      and (p_type is null or w.type = p_type)
  ),
  visible_workflows as (
    select * from owned
    union all
    select * from shared
  )
  select
    vw.id,
    vw.user_id,
    vw.title,
    vw.type,
    vw.prompt_md,
    vw.columns_config,
    vw.practice,
    vw.is_system,
    vw.created_at,
    vw.allow_edit,
    vw.is_owner,
    vw.shared_by_name
  from visible_workflows vw
  order by vw.sort_bucket asc, vw.created_at desc;
$$;

-- ---------------------------------------------------------------------------
-- Assistant chats
-- ---------------------------------------------------------------------------

create table if not exists public.chats (
  id uuid primary key default gen_random_uuid(),
  project_id uuid references public.projects(id) on delete cascade,
  user_id text not null,
  title text,
  created_at timestamptz not null default now()
);

create index if not exists idx_chats_user
  on public.chats(user_id);

create index if not exists idx_chats_project
  on public.chats(project_id);

create or replace function public.get_chats_overview(
  p_user_id text,
  p_limit integer default null
)
returns table (
  id uuid,
  project_id uuid,
  user_id text,
  title text,
  created_at timestamptz
)
language sql
stable
as $$
  select
    c.id,
    c.project_id,
    c.user_id,
    c.title,
    c.created_at
  from public.chats c
  where c.user_id = p_user_id
     or exists (
      select 1
      from public.projects p
      where p.id = c.project_id
        and p.user_id = p_user_id
    )
  order by c.created_at desc
  limit case
    when p_limit is null then null
    else greatest(1, least(p_limit, 100))
  end;
$$;

create table if not exists public.chat_messages (
  id uuid primary key default gen_random_uuid(),
  chat_id uuid not null references public.chats(id) on delete cascade,
  role text not null,
  content jsonb,
  files jsonb,
  annotations jsonb,
  created_at timestamptz not null default now()
);

create index if not exists idx_chat_messages_chat
  on public.chat_messages(chat_id);

do $$
begin
  if not exists (
    select 1
    from pg_constraint
    where conname = 'document_edits_chat_message_id_fkey'
      and conrelid = 'public.document_edits'::regclass
  ) then
    alter table public.document_edits
      add constraint document_edits_chat_message_id_fkey
      foreign key (chat_message_id)
      references public.chat_messages(id)
      on delete set null;
  end if;
end;
$$;

-- ---------------------------------------------------------------------------
-- Tabular reviews
-- ---------------------------------------------------------------------------

create table if not exists public.tabular_reviews (
  id uuid primary key default gen_random_uuid(),
  project_id uuid references public.projects(id) on delete cascade,
  user_id text not null,
  title text,
  columns_config jsonb,
  document_ids jsonb,
  workflow_id uuid references public.workflows(id) on delete set null,
  practice text,
  shared_with jsonb not null default '[]'::jsonb,
  created_at timestamptz not null default now(),
  updated_at timestamptz not null default now()
);

create index if not exists idx_tabular_reviews_user
  on public.tabular_reviews(user_id);

create index if not exists idx_tabular_reviews_project
  on public.tabular_reviews(project_id);

create index if not exists tabular_reviews_shared_with_idx
  on public.tabular_reviews using gin (shared_with);

create or replace function public.get_projects_overview(
  p_user_id text,
  p_user_email text default null
)
returns table (
  id uuid,
  user_id text,
  name text,
  cm_number text,
  shared_with jsonb,
  created_at timestamptz,
  updated_at timestamptz,
  is_owner boolean,
  owner_display_name text,
  owner_email text,
  document_count integer,
  chat_count integer,
  review_count integer
)
language sql
stable
as $$
  with visible_projects as (
    select p.*
    from public.projects p
    where p.user_id = p_user_id
       or (
        coalesce(p_user_email, '') <> ''
        and p.user_id <> p_user_id
        and p.shared_with @> jsonb_build_array(p_user_email)
      )
  ),
  document_counts as (
    select d.project_id, count(*)::integer as document_count
    from public.documents d
    where d.project_id in (select vp.id from visible_projects vp)
    group by d.project_id
  ),
  chat_counts as (
    select c.project_id, count(*)::integer as chat_count
    from public.chats c
    where c.project_id in (select vp.id from visible_projects vp)
    group by c.project_id
  ),
  review_counts as (
    select tr.project_id, count(*)::integer as review_count
    from public.tabular_reviews tr
    where tr.project_id in (select vp.id from visible_projects vp)
    group by tr.project_id
  )
  select
    vp.id,
    vp.user_id,
    vp.name,
    vp.cm_number,
    vp.shared_with,
    vp.created_at,
    vp.updated_at,
    vp.user_id = p_user_id as is_owner,
    nullif(trim(up.display_name), '') as owner_display_name,
    null::text as owner_email,
    coalesce(dc.document_count, 0) as document_count,
    coalesce(cc.chat_count, 0) as chat_count,
    coalesce(rc.review_count, 0) as review_count
  from visible_projects vp
  left join public.user_profiles up
    on up.user_id::text = vp.user_id
  left join document_counts dc
    on dc.project_id = vp.id
  left join chat_counts cc
    on cc.project_id = vp.id
  left join review_counts rc
    on rc.project_id = vp.id
  order by vp.created_at desc;
$$;

create table if not exists public.tabular_cells (
  id uuid primary key default gen_random_uuid(),
  review_id uuid not null references public.tabular_reviews(id) on delete cascade,
  document_id uuid not null references public.documents(id) on delete cascade,
  column_index integer not null,
  content text,
  citations jsonb,
  status text not null default 'pending',
  created_at timestamptz not null default now()
);

create index if not exists idx_tabular_cells_review
  on public.tabular_cells(review_id, document_id, column_index);

create or replace function public.get_tabular_reviews_overview(
  p_user_id text,
  p_user_email text default null,
  p_project_id text default null
)
returns table (
  id uuid,
  project_id uuid,
  user_id text,
  title text,
  columns_config jsonb,
  document_ids jsonb,
  workflow_id uuid,
  shared_with jsonb,
  created_at timestamptz,
  updated_at timestamptz,
  is_owner boolean,
  document_count integer
)
language sql
stable
as $$
  with accessible_projects as (
    select p.id
    from public.projects p
    where p.user_id = p_user_id
       or (
        coalesce(p_user_email, '') <> ''
        and p.user_id <> p_user_id
        and p.shared_with @> jsonb_build_array(p_user_email)
      )
  ),
  visible_reviews as (
    select tr.*
    from public.tabular_reviews tr
    where (p_project_id is null or tr.project_id::text = p_project_id)
      and (
        p_project_id is null
        or exists (
          select 1
          from accessible_projects ap
          where ap.id::text = p_project_id
        )
      )
      and (
        tr.user_id = p_user_id
        or (
          tr.project_id in (select ap.id from accessible_projects ap)
          and tr.user_id <> p_user_id
        )
        or (
          p_project_id is null
          and coalesce(p_user_email, '') <> ''
          and tr.user_id <> p_user_id
          and tr.shared_with @> jsonb_build_array(p_user_email)
        )
      )
  ),
  cell_document_counts as (
    select
      tc.review_id,
      count(distinct tc.document_id)::integer as document_count
    from public.tabular_cells tc
    where tc.review_id in (select vr.id from visible_reviews vr)
    group by tc.review_id
  )
  select
    vr.id,
    vr.project_id,
    vr.user_id,
    vr.title,
    vr.columns_config,
    vr.document_ids,
    vr.workflow_id,
    vr.shared_with,
    vr.created_at,
    vr.updated_at,
    vr.user_id = p_user_id as is_owner,
    case
      when jsonb_typeof(vr.document_ids) = 'array'
        then (
          select count(distinct doc_id.value)::integer
          from jsonb_array_elements_text(vr.document_ids) as doc_id(value)
        )
      else coalesce(cdc.document_count, 0)
    end as document_count
  from visible_reviews vr
  left join cell_document_counts cdc
    on cdc.review_id = vr.id
  order by vr.created_at desc;
$$;

create table if not exists public.tabular_review_chats (
  id uuid primary key default gen_random_uuid(),
  review_id uuid not null references public.tabular_reviews(id) on delete cascade,
  user_id text not null,
  title text,
  created_at timestamptz not null default now(),
  updated_at timestamptz not null default now()
);

create index if not exists tabular_review_chats_review_idx
  on public.tabular_review_chats(review_id, updated_at desc);

create index if not exists tabular_review_chats_user_idx
  on public.tabular_review_chats(user_id);

create table if not exists public.tabular_review_chat_messages (
  id uuid primary key default gen_random_uuid(),
  chat_id uuid not null references public.tabular_review_chats(id) on delete cascade,
  role text not null,
  content jsonb,
  annotations jsonb,
  created_at timestamptz not null default now()
);

create index if not exists tabular_review_chat_messages_chat_idx
  on public.tabular_review_chat_messages(chat_id, created_at);

-- ---------------------------------------------------------------------------
-- CourtListener bulk-data indexes
-- ---------------------------------------------------------------------------

create table if not exists public.courtlistener_citation_index (
  id bigint primary key,
  volume text not null,
  reporter text not null,
  page text not null,
  type integer,
  cluster_id bigint not null,
  date_created timestamptz,
  date_modified timestamptz
);

create index if not exists courtlistener_citation_lookup_idx
  on public.courtlistener_citation_index(volume, reporter, page);

create index if not exists courtlistener_citation_cluster_idx
  on public.courtlistener_citation_index(cluster_id);

alter table public.courtlistener_citation_index enable row level security;

create table if not exists public.courtlistener_opinion_cluster_index (
  id bigint primary key,
  case_name text,
  case_name_short text,
  case_name_full text,
  slug text,
  date_filed date,
  citation_count integer,
  precedential_status text,
  filepath_pdf_harvard text,
  filepath_json_harvard text,
  docket_id bigint
);

alter table public.courtlistener_opinion_cluster_index enable row level security;

-- ---------------------------------------------------------------------------
-- Direct client grant hardening
-- ---------------------------------------------------------------------------
--
-- The frontend uses Supabase directly only for authentication. Application
-- data access goes through the backend API with the service role after the
-- backend verifies the user's JWT. Do not grant the browser anon/authenticated
-- roles direct table privileges for backend-owned data.

revoke all on public.user_profiles from anon, authenticated;
revoke all on public.projects from anon, authenticated;
revoke all on public.project_subfolders from anon, authenticated;
revoke all on public.documents from anon, authenticated;
revoke all on public.document_versions from anon, authenticated;
revoke all on public.document_edits from anon, authenticated;
revoke all on public.workflows from anon, authenticated;
revoke all on public.hidden_workflows from anon, authenticated;
revoke all on public.workflow_shares from anon, authenticated;
revoke all on public.chats from anon, authenticated;
revoke all on public.chat_messages from anon, authenticated;
revoke all on public.tabular_reviews from anon, authenticated;
revoke all on public.tabular_cells from anon, authenticated;
revoke all on public.tabular_review_chats from anon, authenticated;
revoke all on public.tabular_review_chat_messages from anon, authenticated;
revoke all on public.user_api_keys from anon, authenticated;
revoke all on public.user_mcp_connectors from anon, authenticated;
revoke all on public.user_mcp_oauth_tokens from anon, authenticated;
revoke all on public.user_mcp_oauth_states from anon, authenticated;
revoke all on public.user_mcp_connector_tools from anon, authenticated;
revoke all on public.user_mcp_tool_audit_logs from anon, authenticated;
revoke all on public.courtlistener_citation_index from anon, authenticated;
revoke all on public.courtlistener_opinion_cluster_index from anon, authenticated;