add schemas and swap browser_use_runner in docker with browser and now agent use skill from the other container

hermes_code/package.json

@@ -4,7 +4,7 @@
   "description": "An AI agent with advanced tool-calling capabilities, featuring a flexible toolsets system for organizing and managing tools.",
   "private": true,
   "scripts": {
-    "postinstall": "echo '✅ Browser tools ready. Run: python run_agent.py --help'"
+    "postinstall": "echo '✅ Hermes Node hooks ready. Run: python run_agent.py --help'"
   },
   "repository": {
     "type": "git",
@@ -15,9 +15,7 @@
     "url": "https://github.com/NousResearch/Hermes-Agent/issues"
   },
   "homepage": "https://github.com/NousResearch/Hermes-Agent#readme",
-  "dependencies": {
-    "agent-browser": "^0.13.0"
-  },
+  "dependencies": {},
   "engines": {
     "node": ">=18.0.0"
   }

hermes_code/pyproject.toml

@@ -34,10 +34,6 @@ dependencies = [
   "faster-whisper>=1.0.0,<2",
   # Skills Hub (GitHub App JWT auth — optional, only needed for bot identity)
   "PyJWT[crypto]>=2.10.1,<3",
-  "browser-use>=0.12.5",
-  "playwright>=1.49.0",
-  "playwright-stealth>=1.0.6",
-  "langchain-openai>=1.1.12",
 ]
 
 [project.optional-dependencies]

hermes_code/requirements.txt

@@ -19,9 +19,6 @@ PyJWT[crypto]
 # Web tools
 firecrawl-py
 parallel-web>=0.4.2
-browser-use>=0.12.5
-playwright
-playwright-stealth
 
 # Image generation
 fal-client
@@ -36,5 +33,4 @@ croniter
 python-telegram-bot>=20.0
 discord.py>=2.0
 aiohttp>=3.9.0
-langchain-openai>=1.1.12,
 

hermes_code/scripts/install.sh

@@ -874,41 +874,14 @@ install_node_deps() {
     fi
 
     if [ -f "$INSTALL_DIR/package.json" ]; then
-        log_info "Installing Node.js dependencies (browser tools)..."
+        log_info "Installing Node.js dependencies..."
         cd "$INSTALL_DIR"
         npm install --silent 2>/dev/null || {
-            log_warn "npm install failed (browser tools may not work)"
+            log_warn "npm install failed (some Node.js helpers may not work)"
         }
         log_success "Node.js dependencies installed"
 
-        # Install Playwright browser + system dependencies.
-        # Playwright's install-deps only supports apt/dnf/zypper natively.
-        # For Arch/Manjaro we install the system libs via pacman first.
-        log_info "Installing browser engine (Playwright Chromium)..."
-        case "$DISTRO" in
-            arch|manjaro)
-                if command -v pacman &> /dev/null; then
-                    log_info "Arch/Manjaro detected — installing Chromium system dependencies via pacman..."
-                    if command -v sudo &> /dev/null && sudo -n true 2>/dev/null; then
-                        sudo NEEDRESTART_MODE=a pacman -S --noconfirm --needed \
-                            nss atk at-spi2-core cups libdrm libxkbcommon mesa pango cairo alsa-lib >/dev/null 2>&1 || true
-                    elif [ "$(id -u)" -eq 0 ]; then
-                        pacman -S --noconfirm --needed \
-                            nss atk at-spi2-core cups libdrm libxkbcommon mesa pango cairo alsa-lib >/dev/null 2>&1 || true
-                    else
-                        log_warn "Cannot install browser deps without sudo. Run manually:"
-                        log_warn "  sudo pacman -S nss atk at-spi2-core cups libdrm libxkbcommon mesa pango cairo alsa-lib"
-                    fi
-                fi
-                cd "$INSTALL_DIR" && npx playwright install chromium 2>/dev/null || true
-                ;;
-            *)
-                log_info "Playwright may request sudo to install browser system dependencies (shared libraries)."
-                log_info "This is standard Playwright setup — Hermes itself does not require root access."
-                cd "$INSTALL_DIR" && npx playwright install --with-deps chromium 2>/dev/null || true
-                ;;
-        esac
-        log_success "Browser engine installed"
+        log_info "Skipping local Playwright/Chromium bootstrap (browser automation runs in browser container)"
     fi
 
     # Install WhatsApp bridge dependencies

hermes_code/tools/browser_use_tool.py

@@ -1,61 +1,40 @@
 import json
 import os
-import asyncio
-import socket
-from browser_use import Agent, Browser, ChatOpenAI
+from urllib import error, request
 from tools.registry import registry
 
 
-async def run_browser_task(task):
-    browser_host = "browser"
-    browser_port = 9222
-    BROWSER_VIEW_URL = os.getenv("BROWSER_VIEW_URL", "")
-    
-    try:
-        browser_ip = socket.gethostbyname(browser_host)
-        cdp_url = f"http://{browser_ip}:{browser_port}"
-    except Exception:
-        cdp_url = f"http://{browser_host}:{browser_port}"
+def run_browser_task(task):
+    if not task or not str(task).strip():
+        return json.dumps({"success": False, "error": "Task is required"}, ensure_ascii=False)
 
-    browser = Browser(cdp_url=cdp_url)
-
-    llm = ChatOpenAI(
-        model=os.getenv("MODEL_DEFAULT", "qwen3.5-122b"),
-        api_key=os.getenv("OPENAI_API_KEY"),
-        base_url=os.getenv("OPENAI_BASE_URL"),
-        temperature=0.0,
-    )
-
-    agent = Agent(
-        task=task,
-        llm=llm,
-        browser=browser,
-        use_vision=False
-    )
+    rpc_url = os.getenv("BROWSER_USE_RPC_URL", "http://browser:8787/run")
+    timeout_sec = int(os.getenv("BROWSER_USE_RPC_TIMEOUT", "900"))
+    payload = json.dumps({"task": task}).encode("utf-8")
+    req = request.Request(rpc_url, data=payload, headers={"Content-Type": "application/json"}, method="POST")
 
     try:
-        history = await agent.run()
-        final_result = history.final_result()
-
-        response = {
-            "success": True,
-            "result": final_result,
-            "browser_view": BROWSER_VIEW_URL
-        }
-        return json.dumps(response, ensure_ascii=False)
-        
-    except Exception as e:
-        return json.dumps({
-            "success": False, 
-            "error": f"Browser automation failed: {str(e)}"
-        }, ensure_ascii=False)
-        
-    finally:
-        if browser:
-            try:
-                await browser.close()
-            except Exception:
-                pass
+        with request.urlopen(req, timeout=timeout_sec) as resp:
+            body = resp.read().decode("utf-8")
+            return body
+    except error.HTTPError as http_err:
+        body = http_err.read().decode("utf-8", errors="replace")
+        return json.dumps(
+            {
+                "success": False,
+                "error": f"browser-use RPC returned HTTP {http_err.code}",
+                "details": body,
+            },
+            ensure_ascii=False,
+        )
+    except Exception as err:
+        return json.dumps(
+            {
+                "success": False,
+                "error": f"browser-use RPC request failed: {err}",
+            },
+            ensure_ascii=False,
+        )
 
 
 registry.register(
@@ -81,6 +60,6 @@ registry.register(
         }
     },
  
-    handler=lambda args, **kw: asyncio.run(run_browser_task(args.get("task"))),
+    handler=lambda args, **kw: run_browser_task(args.get("task")),
     emoji="🌐",
 )