wcgw 1.5.3__py3-none-any.whl → 1.5.4__py3-none-any.whl

wcgw/client/anthropic_client.py

@@ -131,6 +131,7 @@ def loop(
     first_message: Optional[str] = None,
     limit: Optional[float] = None,
     resume: Optional[str] = None,
+    computer_use: bool = False,
 ) -> tuple[str, float]:
     load_dotenv()
 
@@ -222,10 +223,14 @@ def loop(
 - Use SEARCH/REPLACE blocks to edit the file.
 """,
         ),
-        ToolParam(
-            input_schema=GetScreenInfo.model_json_schema(),
-            name="GetScreenInfo",
-            description="""
+    ]
+
+    if computer_use:
+        tools += [
+            ToolParam(
+                input_schema=GetScreenInfo.model_json_schema(),
+                name="GetScreenInfo",
+                description="""
 - Important: call this first in the conversation before ScreenShot, Mouse, and Keyboard tools.
 - Get display information of a linux os running on docker using image "ghcr.io/anthropics/anthropic-quickstarts:computer-use-demo-latest"
 - If user hasn't provided docker image id, check using `docker ps` and provide the id.
@@ -233,11 +238,11 @@ def loop(
 - Connects shell to the docker environment.
 - Note: once this is called, the shell enters the docker environment. All bash commands will run over there.
 """,
-        ),
-        ToolParam(
-            input_schema=ScreenShot.model_json_schema(),
-            name="ScreenShot",
-            description="""
+            ),
+            ToolParam(
+                input_schema=ScreenShot.model_json_schema(),
+                name="ScreenShot",
+                description="""
 - Capture screenshot of the linux os on docker.
 - All actions on UI using mouse and keyboard return within 0.5 seconds.
     * So if you're doing something that takes longer for UI to update like heavy page loading, keep checking UI for update usign ScreenShot upto 10 turns. 
@@ -246,27 +251,27 @@ def loop(
     * If you don't notice even slightest of the change, it's likely you clicked on the wrong place.
 
 """,
-        ),
-        ToolParam(
-            input_schema=Mouse.model_json_schema(),
-            name="Mouse",
-            description="""
+            ),
+            ToolParam(
+                input_schema=Mouse.model_json_schema(),
+                name="Mouse",
+                description="""
 - Interact with the linux os on docker using mouse.
 - Uses xdotool
 - About left_click_drag: the current mouse position will be used as the starting point, click and drag to the given x, y coordinates. Useful in things like sliders, moving things around, etc.
 """,
-        ),
-        ToolParam(
-            input_schema=Keyboard.model_json_schema(),
-            name="Keyboard",
-            description="""
+            ),
+            ToolParam(
+                input_schema=Keyboard.model_json_schema(),
+                name="Keyboard",
+                description="""
 - Interact with the linux os on docker using keyboard.
 - Emulate keyboard input to the screen
 - Uses xdootool to send keyboard input, keys like Return, BackSpace, Escape, Page_Up, etc. can be used.
 - Do not use it to interact with Bash tool.
 """,
-        ),
-    ]
+            ),
+        ]
     uname_sysname = os.uname().sysname
     uname_machine = os.uname().machine
 

wcgw/client/cli.py

@@ -16,6 +16,7 @@ def loop(
     first_message: Optional[str] = None,
     limit: Optional[float] = None,
     resume: Optional[str] = None,
+    computer_use: bool = False,
     version: bool = typer.Option(False, "--version", "-v"),
 ) -> tuple[str, float]:
     if version:
@@ -27,6 +28,7 @@ def loop(
             first_message=first_message,
             limit=limit,
             resume=resume,
+            computer_use=computer_use,
         )
     else:
         return openai_loop(

wcgw/client/mcp_server/Readme.md

@@ -31,15 +31,39 @@ Then restart claude app.
 
 ### [Optional] Computer use support using desktop on docker
 
-Computer use is enabled by default. Claude will be able to connect to any docker container with linux environment. Native system control isn't supported outside docker.
+Computer use is disabled by default. Add `--computer-use` to enable it. This will add necessary tools to Claude including ScreenShot, Mouse and Keyboard control.
 
-First run a sample docker image with desktop and optionally VNC connection:
+```json
+{
+  "mcpServers": {
+    "wcgw": {
+      "command": "uv",
+      "args": [
+        "tool",
+        "run",
+        "--from",
+        "wcgw@latest",
+        "--python",
+        "3.12",
+        "wcgw_mcp",
+        "--computer-use"
+      ]
+    }
+  }
+}
+```
+
+Claude will be able to connect to any docker container with linux environment. Native system control isn't supported outside docker.
+
+You'll need to run a docker image with desktop and optional VNC connection. Here's a demo image:
 
 ```sh
 docker run -p 6080:6080 ghcr.io/anthropics/anthropic-quickstarts:computer-use-demo-latest
 ```
 
-Connect to `http://localhost:6080/vnc.html` for desktop view (VNC) of the system running in the docker. Then ask claude desktop app to control the docker os.
+Then ask claude desktop app to control the docker os. It'll connect to the docker container and control it.
+
+Connect to `http://localhost:6080/vnc.html` for desktop view (VNC) of the system running in the docker.
 
 ## Usage
 

wcgw/client/mcp_server/__init__.py

@@ -1,10 +1,14 @@
 from wcgw.client.mcp_server import server
 import asyncio
+from typer import Typer
 
+main = Typer()
 
-def main():
+
+@main.command()
+def app(computer_use: bool = False) -> None:
     """Main entry point for the package."""
-    asyncio.run(server.main())
+    asyncio.run(server.main(computer_use))
 
 
 # Optionally expose other important items at package level

wcgw/client/mcp_server/server.py

@@ -32,6 +32,8 @@ from ..computer_use import SLEEP_TIME_MAX_S
 
 tools.TIMEOUT = SLEEP_TIME_MAX_S
 
+COMPUTER_USE_ON_DOCKER_ENABLED = False
+
 server = Server("wcgw")
 
 
@@ -71,7 +73,7 @@ async def handle_list_tools() -> list[types.Tool]:
     ) as f:
         diffinstructions = f.read()
 
-    return [
+    tools = [
         ToolParam(
             inputSchema=Initialize.model_json_schema(),
             name="Initialize",
@@ -142,17 +144,13 @@ async def handle_list_tools() -> list[types.Tool]:
 """
             + diffinstructions,
         ),
-        ToolParam(
-            inputSchema=ReadImage.model_json_schema(),
-            name="ReadImage",
-            description="""
-- Read an image from the shell.
-""",
-        ),
-        ToolParam(
-            inputSchema=GetScreenInfo.model_json_schema(),
-            name="GetScreenInfo",
-            description="""
+    ]
+    if COMPUTER_USE_ON_DOCKER_ENABLED:
+        tools += [
+            ToolParam(
+                inputSchema=GetScreenInfo.model_json_schema(),
+                name="GetScreenInfo",
+                description="""
 - Important: call this first in the conversation before ScreenShot, Mouse, and Keyboard tools.
 - Get display information of a linux os running on docker using image "ghcr.io/anthropics/anthropic-quickstarts:computer-use-demo-latest"
 - If user hasn't provided docker image id, check using `docker ps` and provide the id.
@@ -160,11 +158,11 @@ async def handle_list_tools() -> list[types.Tool]:
 - Connects shell to the docker environment.
 - Note: once this is called, the shell enters the docker environment. All bash commands will run over there.
 """,
-        ),
-        ToolParam(
-            inputSchema=ScreenShot.model_json_schema(),
-            name="ScreenShot",
-            description="""
+            ),
+            ToolParam(
+                inputSchema=ScreenShot.model_json_schema(),
+                name="ScreenShot",
+                description="""
 - Capture screenshot of the linux os on docker.
 - All actions on UI using mouse and keyboard return within 0.5 seconds.
     * So if you're doing something that takes longer for UI to update like heavy page loading, keep checking UI for update usign ScreenShot upto 10 turns. 
@@ -172,28 +170,30 @@ async def handle_list_tools() -> list[types.Tool]:
     * After 10 turns of no change, ask user for permission to keep checking.
     * If you don't notice even slightest of the change, it's likely you clicked on the wrong place.
 """,
-        ),
-        ToolParam(
-            inputSchema=Mouse.model_json_schema(),
-            name="Mouse",
-            description="""
+            ),
+            ToolParam(
+                inputSchema=Mouse.model_json_schema(),
+                name="Mouse",
+                description="""
 - Interact with the linux os on docker using mouse.
 - Uses xdotool
 - About left_click_drag: the current mouse position will be used as the starting point, click and drag to the given x, y coordinates. Useful in things like sliders, moving things around, etc.
 """,
-        ),
-        ToolParam(
-            inputSchema=Keyboard.model_json_schema(),
-            name="Keyboard",
-            description="""
+            ),
+            ToolParam(
+                inputSchema=Keyboard.model_json_schema(),
+                name="Keyboard",
+                description="""
 - Interact with the linux os on docker using keyboard.
 - Emulate keyboard input to the screen
 - Uses xdootool to send keyboard input, keys like Return, BackSpace, Escape, Page_Up, etc. can be used.
 - Do not use it to interact with Bash tool.
 - Make sure you've selected a text area or an editable element before sending text.
 """,
-        ),
-    ]
+            ),
+        ]
+
+    return tools
 
 
 @server.call_tool()  # type: ignore
@@ -263,7 +263,11 @@ async def handle_call_tool(
     return content
 
 
-async def main() -> None:
+async def main(computer_use: bool) -> None:
+    global COMPUTER_USE_ON_DOCKER_ENABLED
+    if computer_use:
+        COMPUTER_USE_ON_DOCKER_ENABLED = True
+
     version = importlib.metadata.version("wcgw")
     # Run the server using stdin/stdout streams
     async with mcp.server.stdio.stdio_server() as (read_stream, write_stream):

wcgw/client/openai_client.py

@@ -123,6 +123,7 @@ def loop(
     first_message: Optional[str] = None,
     limit: Optional[float] = None,
     resume: Optional[str] = None,
+    computer_use: bool = False,
 ) -> tuple[str, float]:
     load_dotenv()
 

{wcgw-1.5.3.dist-info → wcgw-1.5.4.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: wcgw
-Version: 1.5.3
+Version: 1.5.4
 Summary: What could go wrong giving full shell access to chatgpt?
 Project-URL: Homepage, https://github.com/rusiaaman/wcgw
 Author-email: Aman Rusia <gapypi@arcfu.com>

{wcgw-1.5.3.dist-info → wcgw-1.5.4.dist-info}/RECORD

@@ -2,21 +2,21 @@ wcgw/__init__.py,sha256=9K2QW7QuSLhMTVbKbBYd9UUp-ZyrfBrxcjuD_xk458k,118
 wcgw/types_.py,sha256=rDz4olJS2zvYC13jzeOppA2tci-tVDyWAqeA5BesAaU,1773
 wcgw/client/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 wcgw/client/__main__.py,sha256=wcCrL4PjG51r5wVKqJhcoJPTLfHW0wNbD31DrUN0MWI,28
-wcgw/client/anthropic_client.py,sha256=c6nNNojsWLFWjMJcEsEp7g31Ps9SIgCqMk4dhrOn9V4,18314
-wcgw/client/cli.py,sha256=Oja42CHkVO8puqOXflko9NeephYCMa85aBmQTEjBZtI,932
+wcgw/client/anthropic_client.py,sha256=r-8uv1qmORmqCWCeCrWS14589kn1qz7cq4r2TI3SgFQ,18477
+wcgw/client/cli.py,sha256=-z0kpDAW3mzfQrQeZfaVJhBCAQY3HXnt9GdgQ8s-u0Y,1003
 wcgw/client/common.py,sha256=grH-yV_4tnTQZ29xExn4YicGLxEq98z-HkEZwH0ReSg,1410
 wcgw/client/computer_use.py,sha256=eGiINKfgY8WWT-NDUa6vUKd1MTWE7dTjSlvjZHPCWzc,14870
 wcgw/client/diff-instructions.txt,sha256=s5AJKG23JsjwRYhFZFQVvwDpF67vElawrmdXwvukR1A,1683
-wcgw/client/openai_client.py,sha256=L61ajFVQW2QPS3C0n1YsjgF4vQKfMIZHmp6iFBHutX8,17748
+wcgw/client/openai_client.py,sha256=F5TEv5DhU9twsywSZGtuVkPo6xVaaoaEjvIh88FnIUQ,17780
 wcgw/client/openai_utils.py,sha256=YNwCsA-Wqq7jWrxP0rfQmBTb1dI0s7dWXzQqyTzOZT4,2629
 wcgw/client/sys_utils.py,sha256=GajPntKhaTUMn6EOmopENWZNR2G_BJyuVbuot0x6veI,1376
 wcgw/client/tools.py,sha256=d7Fni7JU3aOh2vXBAw5k5rsxkdQVcxoxc5vipvEsA2g,32680
-wcgw/client/mcp_server/Readme.md,sha256=1hNZtqltsORug7OzUjjoK5O8q5s9-Y3S0_rlzT-Wfg4,2033
-wcgw/client/mcp_server/__init__.py,sha256=cQ7PUrEmXUpio8x0SEoGWP5hCRPd7z2bAkNCbYbtTys,236
-wcgw/client/mcp_server/server.py,sha256=i1nn16LSsbySm3LCt_T_D2G33nPtGNDJakp5FzKV6vQ,10416
+wcgw/client/mcp_server/Readme.md,sha256=I8N4dHkTUVGNQ63BQkBMBhCCBTgqGOSF_pUR6iOEiUk,2495
+wcgw/client/mcp_server/__init__.py,sha256=ucevqn5nkVG4ReENdXr_4ErpcTxj_4ZQ5H27VlCJzmI,337
+wcgw/client/mcp_server/server.py,sha256=M9pJ3DktGsxf6cufXbZ0xxs0HIKNLGc75O_biV2UKYA,10571
 wcgw/relay/serve.py,sha256=RUcUeyL4Xt0EEo12Ul6VQjb4tRle4uIdsa85v7XXxEw,8771
 wcgw/relay/static/privacy.txt,sha256=s9qBdbx2SexCpC_z33sg16TptmAwDEehMCLz4L50JLc,529
-wcgw-1.5.3.dist-info/METADATA,sha256=9raP3OP6KWq5bzIEuwJ41_04g5fS5m1HVSNm-YI8Dkw,6508
-wcgw-1.5.3.dist-info/WHEEL,sha256=C2FUgwZgiLbznR-k0b_5k3Ai_1aASOXDss3lzCUsUug,87
-wcgw-1.5.3.dist-info/entry_points.txt,sha256=eKo1omwbAggWlQ0l7GKoR7uV1-j16nk9tK0BhC2Oz_E,120
-wcgw-1.5.3.dist-info/RECORD,,
+wcgw-1.5.4.dist-info/METADATA,sha256=qui2vxPVk00UvBUgtJayNuOP4USMTIivTJ2gdwe4Qps,6508
+wcgw-1.5.4.dist-info/WHEEL,sha256=C2FUgwZgiLbznR-k0b_5k3Ai_1aASOXDss3lzCUsUug,87
+wcgw-1.5.4.dist-info/entry_points.txt,sha256=eKo1omwbAggWlQ0l7GKoR7uV1-j16nk9tK0BhC2Oz_E,120
+wcgw-1.5.4.dist-info/RECORD,,