fix(ai): count a running sd-cli as the sd engine

fanfictioner runs sd-cli one-shot, not through sd-server, so the sd row
read down while a CLI run held VRAM and generated. A running sd-cli now
reads up and busy, its model marked (sd-cli), and outranks the web UI.
The model is the first --diffusion-model/-m: sd-cli's prompt follows and
is free text that may contain "-m".

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

1 files changed, 12 insertions, 0 deletions

diff --git a/desktop/modules/ai/test-ai-state.sh b/desktop/modules/ai/test-ai-state.sh
index bb6038d..e3e8ba0 100755
--- a/desktop/modules/ai/test-ai-state.sh
+++ b/desktop/modules/ai/test-ai-state.sh
@@ -51,6 +51,9 @@ STUB
 cat > "$stub/pgrep" <<'STUB'
 #!/bin/bash
 case "$*" in
+    *sd-cli*)
+        [[ -n "${T_CLI:-}" ]] || exit 1
+        echo "16618 sd-cli --diffusion-model /data/SD/Krea-2-Turbo-Q6_K.gguf --vae /data/SD/vae/wan_2.1_vae.safetensors -p a prompt -m ignored" ;;
     *sd-server*)
         [[ -n "${T_SD:-}" ]] || exit 1
         echo "4242 sd-server --listen-port 7860 --diffusion-model /data/SD/z_image_turbo-Q8_0.gguf --vae /data/SD/vae/flux1-ae.safetensors" ;;
@@ -128,6 +131,15 @@ touch -d '1 minute ago' "$log"
 got="$(run T_SD=1 | sed -n 2p)"
 check "sd running, log quiet" "$(l engine sd up idle z_image_turbo-Q8_0.gguf)" "$got"
 
+# A one-shot sd-cli (fanfictioner's) holds VRAM and is generating for as
+# long as it runs, whatever the web UI is doing. Its prompt is free text, so
+# a "-m" inside it must not be read as the model.
+got="$(run T_CLI=1 | sed -n 2p)"
+check "sd-cli running" "$(l engine sd up busy "Krea-2-Turbo-Q6_K.gguf (sd-cli)")" "$got"
+
+got="$(run T_CLI=1 T_SD=1 | sed -n 2p)"
+check "sd-cli wins over an idle web UI" "$(l engine sd up busy "Krea-2-Turbo-Q6_K.gguf (sd-cli)")" "$got"
+
 want="$(l app assistant running -; l app chat running -; l app imggen running realvis; l app fanfic running 31337)"
 got="$(run T_ASSIST=1 T_CHAT=1 T_IMG=1 T_FANFIC=1 | tail -n4)"
 check "every app running" "$want" "$got"