diff --git a/dankclaw/voiceservice/tts-engine-warm.sh b/dankclaw/voiceservice/tts-engine-warm.sh
index f835ece..e83f4c4 100755
--- a/dankclaw/voiceservice/tts-engine-warm.sh
+++ b/dankclaw/voiceservice/tts-engine-warm.sh
@@ -16,17 +16,17 @@
 #      voice that did not warm becomes a deploy result instead of only a journal
 #      line (the unit's `-` prefix deliberately cannot fail the unit).
 #
-# WHY IT EXISTS AT ALL (the cold-voice law). The FIRST speech request for a
-# registered voice decodes the whole reference clip to snapshot its codec state:
-# 6.2-8.0 s, first audio byte at 5,530 ms. Every later request for that voice
-# restores the snapshot in 0.5 ms. Two facts make that a boot problem rather than
-# a curiosity: the voice registry is MEMORY ONLY (a restart forgets every voice),
-# and re-registering a voice does not warm it — only synthesizing does. So a boot
-# that registers the voices and stops leaves the first caller of the day paying
-# 5.5 s of silence. This script speaks one short sentence per voice instead, then
-# speaks a second one so the journal carries the warm number too: the line
-# `voice <name> cold=…ms warm=…ms` is the repo's standing proof that the second
-# request of a voice answers well under a second.
+# WHY IT EXISTS AT ALL (the registry law). The voice registry is MEMORY ONLY: a
+# restart forgets every voice, and re-registering a voice does not warm it — only
+# synthesizing does. A boot that registers nothing therefore leaves frank-tts
+# answering 400 for every name, so this script registers every pack it finds. The
+# per-voice cold penalty is real but small, and smaller than the trial reported:
+# measured on the deployed build 2026-10-18 against an empty registry, the FIRST
+# synthesis of a freshly registered voice cost 1,883 ms against 1,226-1,283 ms for
+# the requests after it — about 0.6 s, not the trial's 6.2-8.0 s. Each voice gets
+# two identical requests so the journal carries both numbers, `voice <name>
+# cold=…ms warm=…ms`; cold ≈ warm is the expected shape, and a large gap is what
+# would mean the snapshot path is broken.
 #
 # THE VOICES. /voices — the host directory /home/dank/voices mounted read-only by
 # the unit, the same packs frank-tts mounts for its own name list. A pack is a
@@ -47,7 +47,9 @@
 #
 # NOTHING HERE IS A SECRET, and nothing is logged that could be: the audio bytes
 # are discarded, the synthesized text comes from the env file, and the reference
-# clip is base64'd straight into the request body without ever being echoed.
+# clip is base64'd into a request body that is written to a temporary file and
+# handed to curl as `@file` — never echoed, and never placed on a command line,
+# because of the size law explained at the mktemp below.
 set -euo pipefail
 
 # The engine's own address, from inside its namespace. Loopback is valid HERE:
@@ -103,6 +105,18 @@ fi
 warm_json="${warm_text//\\/\\\\}"
 warm_json="${warm_json//\"/\\\"}"
 
+# The registration body is a FILE, never an argument. The kernel caps a single
+# command-line string at 32 pages — MAX_ARG_STRLEN = 131,072 bytes — and a shipped
+# reference clip base64's to about 1.27 MB, so passing the body as one argument
+# fails with E2BIG before curl is even executed. The first real boot of the unit
+# (2026-10-18, the three shipped packs) proved it end to end: `/usr/bin/curl:
+# Argument list too long` for every voice, reported as the misleading
+# "registration failed (HTTP no response)". `mktemp` lands in the container's
+# /tmp as the unprivileged `tts` user; one file for the whole pass, overwritten
+# per voice, removed on any exit — it is container scratch, not host state.
+body_file="$(mktemp -t frank-tts-warm.XXXXXX)"
+trap 'rm -f "$body_file"' EXIT
+
 fail=0
 log() { printf 'frank-tts-engine-warm: %s\n' "$*"; }
 
@@ -146,9 +160,16 @@ for dir in "$VOICES_DIR"/*/; do
         continue
     fi
 
-    body="$(printf '{"name":"%s","wav_b64":"%s"}' "$name" "$(base64 -w0 "${dir}reference.wav")")"
+    # The audio enters through a redirection rather than a command substitution:
+    # the file is several hundred KB, and both a shell variable and an argument of
+    # that size are what this script used to die on.
+    {
+        printf '{"name":"%s","wav_b64":"' "$name"
+        base64 -w0 "${dir}reference.wav"
+        printf '"}'
+    } > "$body_file"
     if ! code="$(curl -sS --max-time "$REQUEST_TIMEOUT" -o /dev/null -w '%{http_code}' \
-            -H 'Content-Type: application/json' --data-binary "$body" \
+            -H 'Content-Type: application/json' --data-binary "@${body_file}" \
             "${ENGINE_URL}/v1/audio/voices")" || [[ "$code" != "200" ]]; then
         log "voice '${name}': registration failed (HTTP ${code:-no response}) — the shim will answer 400 for this name"
         fail=1
diff --git a/tests/frank_tts_engine.bats b/tests/frank_tts_engine.bats
index 74d9d95..4290b31 100755
--- a/tests/frank_tts_engine.bats
+++ b/tests/frank_tts_engine.bats
@@ -394,6 +394,37 @@ assert {s["response_format"] for s in speech} == {"wav"}, speech
 PY
 }
 
+@test "a reference clip too big for one argument still registers" {
+    # The kernel caps a SINGLE command-line string at 32 pages = 131,072 bytes
+    # (MAX_ARG_STRLEN), and a shipped reference clip base64's to about 1.27 MB —
+    # so a warmer that hands the registration body to curl as an argument dies
+    # before curl runs. That is exactly what the first real boot of 2026-10-18
+    # hit with the real packs: `/usr/bin/curl: Argument list too long`, then the
+    # misleading "registration failed (HTTP no response)" for all three voices.
+    # The stub engine is a real listener and curl is the real binary, so this
+    # test fails for the same kernel reason the host did.
+    start_engine_stub
+    local big="$TEST_TMP/bigvoices/bigvoice"
+    mkdir -p "$big"
+    head -c 300000 /dev/zero | tr '\0' 'A' > "$big/reference.wav"
+
+    run env ENGINE_URL="http://127.0.0.1:$STUB_PORT" VOICES_DIR="$TEST_TMP/bigvoices" \
+        bash "$WARM"
+    [ "$status" -eq 0 ] || { echo "$output"; false; }
+    [[ "$output" == *"voice bigvoice cold="*"ms warm="*"ms"* ]] || { echo "$output"; false; }
+
+    # And the audio that arrived is the audio that was on disk, whole.
+    python3 - "$TEST_TMP/journal.jsonl" "$big/reference.wav" <<'PY'
+import hashlib, json, sys
+entries = [json.loads(l) for l in open(sys.argv[1])]
+regs = [e for e in entries if e["path"] == "/v1/audio/voices"]
+assert len(regs) == 1, entries
+assert regs[0]["name"] == "bigvoice", regs[0]
+want = hashlib.sha256(open(sys.argv[2], "rb").read()).hexdigest()
+assert regs[0]["wav_sha256"] == want, (len(regs[0]["wav_sha256"]), want)
+PY
+}
+
 @test "the warm sentence survives both readers of the env file intact" {
     # scripts/frank-tts-engine.env is read two ways and they disagree about
     # quotes. bash (`set -a; source`, what the seams and a human do) strips the