diff --git a/drizzle.config.ts b/drizzle.config.ts
index 2928e91..e21e9c2 100644
--- a/drizzle.config.ts
+++ b/drizzle.config.ts
@@ -5,7 +5,11 @@ export default defineConfig({
   out: "./drizzle",
   dialect: "postgresql",
   dbCredentials: {
-    url: process.env.DATABASE_URL ?? "postgres://continuum:continuum@localhost:6432/continuum",
+    // Matches scripts/dev-env.sh's DEV_DATABASE_URL. It pointed at 6432 (a
+    // pooler) where nothing listens in this environment, so a bare db:migrate
+    // failed confusingly. Migrations want Postgres directly in any case: see
+    // the note in src/db/index.ts about parameterised queries through pgCat.
+    url: process.env.DATABASE_URL ?? "postgres://continuum:continuum@localhost:5432/continuum",
   },
   verbose: true,
   strict: true,
diff --git a/scripts/dev-env.sh b/scripts/dev-env.sh
index fe1e4f6..1d8f55e 100755
--- a/scripts/dev-env.sh
+++ b/scripts/dev-env.sh
@@ -35,8 +35,10 @@ cd "$REPO_ROOT"
 PG_VERSION=16
 PG_PORT=5432
 VALKEY_PORT=6379
-# Straight at Postgres, not pgCat on 6432: .env.example points at the pooler for
-# production, and neither backend here runs one.
+# Straight at Postgres, not pgCat: .env.example points at the pooler for
+# production, and neither backend here runs one. Keep it that way even where a
+# pooler is reachable - parameterised queries from postgres-js hang through
+# pgCat 1.2.0, and drizzle-kit cannot migrate through it. See src/db/index.ts.
 DEV_DATABASE_URL="postgres://continuum:continuum@localhost:${PG_PORT}/continuum"
 DEV_VALKEY_URL="redis://localhost:${VALKEY_PORT}"
 
diff --git a/scripts/e2e-authz.sh b/scripts/e2e-authz.sh
index 0358e3e..8a63db3 100755
--- a/scripts/e2e-authz.sh
+++ b/scripts/e2e-authz.sh
@@ -50,7 +50,11 @@ new_uuid() {
   fi
 }
 
-psql_q() { PGPASSWORD=continuum psql -h localhost -U continuum -d continuum -tAq -c "$1"; }
+# Uses DATABASE_URL rather than a hardcoded localhost. The backend under test
+# reads that variable, so a psql pointed anywhere else seeds fixtures into a
+# different database than the one being exercised - and the failure message
+# below names DATABASE_URL, which would then be naming a URL nothing tried.
+psql_q() { psql "$DATABASE_URL" -tAq -c "$1"; }
 
 cleanup() {
   local rc=$?
diff --git a/scripts/e2e-edge.sh b/scripts/e2e-edge.sh
index 74210d3..29c4509 100755
--- a/scripts/e2e-edge.sh
+++ b/scripts/e2e-edge.sh
@@ -62,7 +62,11 @@ new_uuid() {
   fi
 }
 
-psql_q() { PGPASSWORD=continuum psql -h localhost -U continuum -d continuum -tAq -c "$1"; }
+# Uses DATABASE_URL rather than a hardcoded localhost. The backend under test
+# reads that variable, so a psql pointed anywhere else seeds fixtures into a
+# different database than the one being exercised - and the failure message
+# below names DATABASE_URL, which would then be naming a URL nothing tried.
+psql_q() { psql "$DATABASE_URL" -tAq -c "$1"; }
 
 # The farm's operator. A real Supabase user id would come from the project;
 # here it only has to be a uuid that the token's `sub` and the org_members row
diff --git a/src/db/index.ts b/src/db/index.ts
index 8ba70e6..5ac2a2f 100644
--- a/src/db/index.ts
+++ b/src/db/index.ts
@@ -2,8 +2,27 @@ import { drizzle } from "drizzle-orm/postgres-js";
 import postgres from "postgres";
 import * as schema from "./schema";
 
-// Points at pgCat/PgBouncer, not Postgres directly, so keep prepared
-// statements disabled and cap the pool per-instance connections low.
+// `prepare: false` because this is meant to sit behind a transaction-mode
+// pooler, where a named statement prepared on one server connection is not
+// there on the next one. `max` is low for the same reason: the pooler
+// multiplexes, so a high per-instance count buys nothing and starves peers.
+//
+// Measured 2026-10-20, and the reason to read this before assuming the
+// deployment works: against pgCat 1.2.0 this client cannot run at all.
+// Through that pooler an unparameterised `select 1` succeeds, and so does
+// `select count(*) from organisations`, but *any* query carrying a bind
+// parameter hangs forever - with `prepare: false`, with `prepare: true`, with
+// `fetch_types: false`, and inside an explicit transaction alike. The same
+// queries straight at Postgres are fine, and psql's `\bind` through the same
+// pool is fine, so it is specific to how postgres-js frames those messages,
+// not to extended protocol in general. Every Drizzle query binds parameters,
+// so there is no subset of this app that works there.
+//
+// So `prepare: false` is necessary but nowhere near sufficient, and running
+// behind pgCat is an unverified claim rather than a tested path. PgBouncer is
+// the pooler this configuration is actually known to suit. Migrations go
+// straight at Postgres regardless - drizzle.config.ts has no way to pass
+// `prepare: false`, and drizzle-kit hung at commit for the same reason.
 const connectionString = process.env.DATABASE_URL;
 if (!connectionString) {
   throw new Error("DATABASE_URL is not set");