Remove Flow agent from stack (extension, server, gateway, monitor tab)

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

Dockerfile

@@ -1,4 +1,4 @@
-# Build the navigator monitor server (GPT / Gemini / Flow live view) from source.
+# Build the navigator monitor server (GPT / Gemini live view) from source.
 FROM golang:1.22-bookworm AS monitor-build
 WORKDIR /src
 COPY monitor/ ./
@@ -29,7 +29,6 @@ USER root
 # Copy extensions into the container
 COPY free-Chatgpt-api/gpt-extension /opt/gpt-extension
 COPY free-gemini-api/gemini-extension /opt/gemini-extension
-COPY flow-agent/Flow-extension /opt/flow-extension
 
 # 3-tab monitor binary (replaces the single-view base monitor)
 COPY --from=monitor-build /monitor /opt/monitor3/monitor
@@ -41,12 +40,6 @@ COPY --from=profilesync-build /profilesync /opt/profilesync/profilesync
 COPY --from=chatgpt-build /agent /opt/chatgpt/agent
 COPY free-Chatgpt-api/config.json /opt/chatgpt/config.json
 COPY --from=gemini-build /free-gemini-api /opt/gemini-srv/free-gemini-api
-COPY flow-agent/ /opt/flow-srv/
-
-# Flow server is Python (FastAPI/uvicorn). Base image has python3; add pip + deps.
-RUN apt-get update && apt-get install -y --no-install-recommends python3-pip && \
-    pip3 install --no-cache-dir --break-system-packages -r /opt/flow-srv/requirements.txt && \
-    apt-get clean && rm -rf /var/lib/apt/lists/*
 
 # Install Chrome for Testing (unbranded). Branded google-chrome-stable (>=128)
 # silently ignores --load-extension, so unpacked extensions never load. CfT is the
@@ -60,9 +53,8 @@ RUN apt-get update && apt-get install -y --no-install-recommends unzip && \
     apt-get clean && rm -rf /var/lib/apt/lists/*
 
 # Set correct permissions and ownership
-RUN rm -rf /opt/flow-extension/_metadata && \
-    chown -R 1000:1000 /opt/gpt-extension /opt/gemini-extension /opt/flow-extension /opt/chrome-linux64 /opt/monitor3 /opt/profilesync /opt/chatgpt /opt/gemini-srv /opt/flow-srv && \
-    chmod -R 755 /opt/gpt-extension /opt/gemini-extension /opt/flow-extension /opt/chrome-linux64 /opt/monitor3 /opt/profilesync /opt/chatgpt /opt/gemini-srv /opt/flow-srv
+RUN chown -R 1000:1000 /opt/gpt-extension /opt/gemini-extension /opt/chrome-linux64 /opt/monitor3 /opt/profilesync /opt/chatgpt /opt/gemini-srv && \
+    chmod -R 755 /opt/gpt-extension /opt/gemini-extension /opt/chrome-linux64 /opt/monitor3 /opt/profilesync /opt/chatgpt /opt/gemini-srv
 
 # Replace the start script with the non-root version
 COPY --chmod=755 start_hf.sh /start.sh

README.md

@@ -27,9 +27,8 @@ automate workflows for ChatGPT, Google Gemini, and Google Flow.
 |-----------|-----------|-----------|--------------|
 | **Gpt Agent Api** | `free-Chatgpt-api/gpt-extension` | `/opt/gpt-extension` | Auto-syncs ChatGPT cookies to a local ChatGPT Free API server |
 | **Gemini Api Agent** | `free-gemini-api/gemini-extension` | `/opt/gemini-extension` | Auto-syncs Google Gemini cookies to a local Free Gemini API server |
-| **Flow Agent** | `flow-agent/Flow-extension` | `/opt/flow-extension` | Automates Google Flow — T2V / V2V / I2V video + T2I / I2I image generation |
 
-All three are Manifest V3 extensions with background service workers. Each subfolder
+Both are Manifest V3 extensions with background service workers. Each subfolder
 also contains the matching backend server.
 
 ---

docker-compose.yml

@@ -1,17 +1,16 @@
-# Runs the whole stack in one container: headless Chrome + 3 extensions + the monitor
-# (which also serves the API gateway) + the three backend servers.
+# Runs the whole stack in one container: headless Chrome + 2 extensions + the monitor
+# (which also serves the API gateway) + the two backend servers.
 #
-# The chrome image (built from ./Dockerfile) now bakes in all three servers and start_hf.sh
+# The chrome image (built from ./Dockerfile) now bakes in both servers and start_hf.sh
 # launches them on localhost alongside Chrome — exactly what the extensions expect (they dial
-# 127.0.0.1:9225/9226/9227). So there are no separate server containers anymore; that would
+# 127.0.0.1:9225/9226). So there are no separate server containers anymore; that would
 # collide on the same ports. Only port 3001 is published.
 #
 # Internal port map (all on the container's 127.0.0.1):
 #   chrome   : 9222 CDP, 9223 CDP-proxy
-#   monitor  : 3001 UI + API gateway (/gpt /gemini /flow)
+#   monitor  : 3001 UI + API gateway (/gpt /gemini)
 #   chatgpt  : 9225 (ws + http API)
 #   gemini   : 8000 http API, 9226 cookie-WS
-#   flow     : 8101 http API, 8100 ext-callback, 9227 cookie-WS
 
 services:
   chrome:
@@ -20,7 +19,7 @@ services:
     ports:
       - "3001:3001"   # monitor UI + API gateway (the only published port)
     environment:
-      API_KEY: ${API_KEY:-}   # gates /gpt /gemini /flow; from .env. Unset = APIs disabled (503)
+      API_KEY: ${API_KEY:-}   # gates /gpt /gemini; from .env. Unset = APIs disabled (503)
     volumes:
       - chrome-profile:/home/chrome/data   # persists cookies/logins across restarts
     restart: unless-stopped

monitor/main.go

@@ -44,7 +44,6 @@ type knownTab struct {
 var knownTabs = []knownTab{
 	{"ChatGPT", "https://chatgpt.com/", "chatgpt.com"},
 	{"Gemini", "https://gemini.google.com/", "gemini.google.com"},
-	{"Flow", "https://labs.google/fx", "labs.google"},
 }
 
 // cdpTarget mirrors the relevant fields of an entry from GET /json.
@@ -487,8 +486,8 @@ func chromeLogHandler(w http.ResponseWriter, r *http.Request) {
 }
 
 // backendLogHandler serves the per-server logs (API-key gated) so we can diagnose
-// the gpt/gemini/flow backends without shelling into the container:
-//   /logs/chatgpt  /logs/gemini  /logs/flow  (also /logs/monitor)
+// the gpt/gemini backends without shelling into the container:
+//   /logs/chatgpt  /logs/gemini  (also /logs/monitor)
 func backendLogHandler(w http.ResponseWriter, r *http.Request) {
 	if !authOK(w, r) {
 		return
@@ -497,12 +496,11 @@ func backendLogHandler(w http.ResponseWriter, r *http.Request) {
 	paths := map[string]string{
 		"chatgpt": "/home/chrome/chatgpt.log",
 		"gemini":  "/home/chrome/gemini.log",
-		"flow":    "/home/chrome/flow.log",
 		"monitor": "/home/chrome/monitor.log",
 	}
 	p, ok := paths[name]
 	if !ok {
-		http.Error(w, "unknown log (use chatgpt|gemini|flow|monitor)", http.StatusNotFound)
+		http.Error(w, "unknown log (use chatgpt|gemini|monitor)", http.StatusNotFound)
 		return
 	}
 	f, err := os.Open(p)
@@ -528,10 +526,10 @@ func statsHandler(w http.ResponseWriter, r *http.Request) {
 }
 
 // ---- API gateway ----------------------------------------------------------
-// The monitor doubles as the single public entry point for the three backend
+// The monitor doubles as the single public entry point for the backend
 // servers that run on localhost inside this same container. HF exposes only one
-// port (3001), so /gpt, /gemini and /flow are reverse-proxied to the servers'
-// loopback HTTP ports. All three are gated by an API key (fail-closed).
+// port (3001), so /gpt and /gemini are reverse-proxied to the servers'
+// loopback HTTP ports. Both are gated by an API key (fail-closed).
 
 var apiKey = os.Getenv("API_KEY")
 
@@ -626,7 +624,7 @@ func main() {
 	// HuggingFace's own auth already gates who can reach it (the owner). The browser
 	// can't attach ?key= when HF serves the private subdomain, so requiring our key
 	// here locks the owner out of their own live view. Only the AI gateway
-	// (/gpt /gemini /flow) and /logs stay API-key-gated for external programmatic use.
+	// (/gpt /gemini) and /logs stay API-key-gated for external programmatic use.
 	http.HandleFunc("/", func(w http.ResponseWriter, r *http.Request) {
 		if r.URL.Path != "/" {
 			http.NotFound(w, r)
@@ -651,9 +649,8 @@ func main() {
 	// API gateway → the three backend servers on localhost (API-key gated).
 	http.HandleFunc("/gpt/", gatewayHandler("/gpt", "http://127.0.0.1:9225"))
 	http.HandleFunc("/gemini/", gatewayHandler("/gemini", "http://127.0.0.1:8000"))
-	http.HandleFunc("/flow/", gatewayHandler("/flow", "http://127.0.0.1:8101"))
 	if apiKey == "" {
-		log.Println("WARNING: API_KEY not set — /gpt /gemini /flow are disabled (503) until it is set")
+		log.Println("WARNING: API_KEY not set — /gpt /gemini are disabled (503) until it is set")
 	}
 
 	addr := ":3001"

start_hf.sh

@@ -106,13 +106,12 @@ touch /home/chrome/chrome.log
 echo "🔌 Starting CDP Proxy..."
 socat TCP-LISTEN:9223,fork,bind=0.0.0.0 TCP:127.0.0.1:9222 &
 
-# 6a. Start the three backend API servers on localhost. The monitor doubles as a
-#     gateway and fronts them on :3001 (/gpt, /gemini, /flow). Each logs separately
+# 6a. Start the backend API servers on localhost. The monitor doubles as a
+#     gateway and fronts them on :3001 (/gpt, /gemini). Each logs separately
 #     and runs in the background so a crash never blocks the boot.
-echo "🧩 Starting backend API servers (gpt / gemini / flow)..."
+echo "🧩 Starting backend API servers (gpt / gemini)..."
 ( cd /opt/chatgpt    && ./agent                                  >> /home/chrome/chatgpt.log 2>&1 & )
 ( cd /opt/gemini-srv && WS_PORT=9226 PORT=8000 ./free-gemini-api >> /home/chrome/gemini.log  2>&1 & )
-( cd /opt/flow-srv   && WS_PORT=9227 HTTP_PORT=8100 python3 -m uvicorn cli.api:app --host 127.0.0.1 --port 8101 >> /home/chrome/flow.log 2>&1 & )
 
 # 6b. Periodic profile backup (safety net if HF kills the container without a clean
 #     SIGTERM). Best-effort every 5 minutes; the shutdown trap does the final one.
@@ -124,17 +123,17 @@ fi
 # 6c. Provide a valid per-user runtime dir. Chrome's mojo/IPC layer and each MV3
 #     extension service worker need this; without it Chrome logs
 #     "XDG_RUNTIME_DIR is invalid or not set" and the extension service workers
-#     intermittently fail to spawn (DidStartWorkerFail) → the gpt/gemini/flow
+#     intermittently fail to spawn (DidStartWorkerFail) → the gpt/gemini
 #     extensions never connect. Purely additive: just creates a dir and exports it.
 export XDG_RUNTIME_DIR=/tmp/runtime-chrome
 mkdir -p "$XDG_RUNTIME_DIR" && chmod 700 "$XDG_RUNTIME_DIR"
 
 # 7. Launch Chrome Headless
 # NOTE: do NOT add --renderer-process-limit=1 / --process-per-site / --no-zygote here.
-# Each of the 3 extensions runs an MV3 service worker (its own renderer) and the chat
-# flow injects via chrome.scripting.executeScript into the service tab. Capping renderers
+# Each extension runs an MV3 service worker (its own renderer) and the chat
+# pipeline injects via chrome.scripting.executeScript into the service tab. Capping renderers
 # starves those workers — they fail to start (DidStartWorkerFail) so the extension never
-# connects, and when one does start, executeScript hangs → the /gpt /gemini /flow APIs
+# connects, and when one does start, executeScript hangs → the /gpt /gemini APIs
 # hang forever. HF cpu-basic has 16 GB (we use ~1–2 GB), so the memory "savings" weren't
 # needed and broke the extensions. Keep Chrome multi-process.
 echo "✨ Launching Chrome Headless..."
@@ -173,8 +172,8 @@ echo "✨ Launching Chrome Headless..."
   --disable-background-timer-throttling \
   --disable-renderer-backgrounding \
   --disable-backgrounding-occluded-windows \
-  --load-extension=/opt/gpt-extension,/opt/gemini-extension,/opt/flow-extension \
-  --disable-extensions-except=/opt/gpt-extension,/opt/gemini-extension,/opt/flow-extension \
+  --load-extension=/opt/gpt-extension,/opt/gemini-extension \
+  --disable-extensions-except=/opt/gpt-extension,/opt/gemini-extension \
   --disable-translate \
   --memory-pressure-off \
   --disable-client-side-phishing-detection \
@@ -193,4 +192,4 @@ echo "✨ Launching Chrome Headless..."
   --disable-site-isolation-trials \
   --js-flags="--max-old-space-size=512" \
   --metrics-recording-only \
-  'about:blank' >> /home/chrome/chrome.log 2>&1
+  'https://chatgpt.com/' >> /home/chrome/chrome.log 2>&1