wcgw 0.0.8__tar.gz → 0.0.10__tar.gz

wcgw-0.0.10/PKG-INFO

@@ -0,0 +1,97 @@
+Metadata-Version: 2.3
+Name: wcgw
+Version: 0.0.10
+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>
+Requires-Python: <3.13,>=3.8
+Requires-Dist: fastapi>=0.115.0
+Requires-Dist: mypy>=1.11.2
+Requires-Dist: openai>=1.46.0
+Requires-Dist: petname>=2.6
+Requires-Dist: pexpect>=4.9.0
+Requires-Dist: pydantic>=2.9.2
+Requires-Dist: pyte>=0.8.2
+Requires-Dist: python-dotenv>=1.0.1
+Requires-Dist: rich>=13.8.1
+Requires-Dist: shell>=1.0.1
+Requires-Dist: tiktoken==0.7.0
+Requires-Dist: toml>=0.10.2
+Requires-Dist: typer>=0.12.5
+Requires-Dist: types-pexpect>=4.9.0.20240806
+Requires-Dist: uvicorn>=0.31.0
+Requires-Dist: websockets>=13.1
+Description-Content-Type: text/markdown
+
+# Enable shell access on chatgpt.com
+A custom gpt on chatgpt web app to interact with your local shell.
+
+### 🚀 Highlights
+- ⚡ **Full Shell Access**: No restrictions, complete control.
+- ⚡ **Create, Execute, Iterate**: Ask the gpt to keep running compiler checks till all errors are fixed, or ask it to keep checking for the status of a long running command till it's done.
+- ⚡ **Interactive Command Handling**: [beta] Supports interactive commands using arrow keys, interrupt, and ansi escape sequences. 
+
+###  🪜 Steps: 
+1. Run the [cli client](https://github.com/rusiaaman/wcgw?tab=readme-ov-file#client) in any directory of choice.
+2. Share the generated id with this GPT: `https://chatgpt.com/g/g-Us0AAXkRh-wcgw-giving-shell-access`
+3. The custom GPT can now run any command on your cli
+
+## Client
+
+### Option 1: using pip
+Supports python >=3.8 and <3.13
+```sh
+$ pip3 install wcgw
+$ wcgw
+```
+
+### Option 2: using uv
+```sh
+$ curl -LsSf https://astral.sh/uv/install.sh | sh
+$ uv tool run --python 3.12 wcgw
+```
+
+This will print a UUID that you need to share with the gpt.
+
+
+## Chat
+Open the following link or search the "wcgw" custom gpt using "Explore GPTs" on chatgpt.com
+
+https://chatgpt.com/g/g-Us0AAXkRh-wcgw-giving-shell-access
+
+Finally, let the chatgpt know your user id in any format. E.g., "user_id=<your uuid>" followed by rest of your instructions.
+
+NOTE: you can resume a broken connection 
+`wcgw --client-uuid $previous_uuid`
+
+# How it works
+Your commands are relayed through a server I've hosted at https://wcgw.arcfu.com. The code for that is at `src/relay/serve.py`. 
+
+Chat gpt sends a request to the relay server using the user id that you share with it. The relay server holds a websocket with the terminal cilent against the user id and acts as a proxy to pass the request.
+
+It's secure in both the directions. Either a malicious actor or a malicious Chatgpt has to correctly guess your UUID for any security breach. 
+
+NOTE: the relay server doesn't store any data. If you don't trust it then you may host the server on your own and create a custom gpt. Create an issue and I'll be happy to share the full instructions and schema I've given in the custom GPT configuration. 
+
+# Showcase
+
+## Create a todo app using react + typescript + vite
+https://chatgpt.com/share/6717d94d-756c-8005-98a6-d021c7b586aa
+
+## Write unit tests for all files in my current repo
+[Todo]
+
+
+# [Optional] Local shell access with openai API key
+
+Add `OPENAI_API_KEY` and `OPENAI_ORG_ID` env variables.
+
+Clone the repo and run to install `wcgw_local` command
+
+`pip install .`
+
+Then run 
+
+`wcgw_local  --limit 0.1` # Cost limit $0.1 
+
+You can now directly write messages or press enter key to open vim for multiline message and text pasting.

wcgw-0.0.10/README.md

@@ -0,0 +1,72 @@
+# Enable shell access on chatgpt.com
+A custom gpt on chatgpt web app to interact with your local shell.
+
+### 🚀 Highlights
+- ⚡ **Full Shell Access**: No restrictions, complete control.
+- ⚡ **Create, Execute, Iterate**: Ask the gpt to keep running compiler checks till all errors are fixed, or ask it to keep checking for the status of a long running command till it's done.
+- ⚡ **Interactive Command Handling**: [beta] Supports interactive commands using arrow keys, interrupt, and ansi escape sequences. 
+
+###  🪜 Steps: 
+1. Run the [cli client](https://github.com/rusiaaman/wcgw?tab=readme-ov-file#client) in any directory of choice.
+2. Share the generated id with this GPT: `https://chatgpt.com/g/g-Us0AAXkRh-wcgw-giving-shell-access`
+3. The custom GPT can now run any command on your cli
+
+## Client
+
+### Option 1: using pip
+Supports python >=3.8 and <3.13
+```sh
+$ pip3 install wcgw
+$ wcgw
+```
+
+### Option 2: using uv
+```sh
+$ curl -LsSf https://astral.sh/uv/install.sh | sh
+$ uv tool run --python 3.12 wcgw
+```
+
+This will print a UUID that you need to share with the gpt.
+
+
+## Chat
+Open the following link or search the "wcgw" custom gpt using "Explore GPTs" on chatgpt.com
+
+https://chatgpt.com/g/g-Us0AAXkRh-wcgw-giving-shell-access
+
+Finally, let the chatgpt know your user id in any format. E.g., "user_id=<your uuid>" followed by rest of your instructions.
+
+NOTE: you can resume a broken connection 
+`wcgw --client-uuid $previous_uuid`
+
+# How it works
+Your commands are relayed through a server I've hosted at https://wcgw.arcfu.com. The code for that is at `src/relay/serve.py`. 
+
+Chat gpt sends a request to the relay server using the user id that you share with it. The relay server holds a websocket with the terminal cilent against the user id and acts as a proxy to pass the request.
+
+It's secure in both the directions. Either a malicious actor or a malicious Chatgpt has to correctly guess your UUID for any security breach. 
+
+NOTE: the relay server doesn't store any data. If you don't trust it then you may host the server on your own and create a custom gpt. Create an issue and I'll be happy to share the full instructions and schema I've given in the custom GPT configuration. 
+
+# Showcase
+
+## Create a todo app using react + typescript + vite
+https://chatgpt.com/share/6717d94d-756c-8005-98a6-d021c7b586aa
+
+## Write unit tests for all files in my current repo
+[Todo]
+
+
+# [Optional] Local shell access with openai API key
+
+Add `OPENAI_API_KEY` and `OPENAI_ORG_ID` env variables.
+
+Clone the repo and run to install `wcgw_local` command
+
+`pip install .`
+
+Then run 
+
+`wcgw_local  --limit 0.1` # Cost limit $0.1 
+
+You can now directly write messages or press enter key to open vim for multiline message and text pasting.

{wcgw-0.0.8 → wcgw-0.0.10}/pyproject.toml

@@ -1,7 +1,7 @@
 [project]
 authors = [{ name = "Aman Rusia", email = "gapypi@arcfu.com" }]
 name = "wcgw"
-version = "0.0.8"
+version = "0.0.10"
 description = "What could go wrong giving full shell access to chatgpt?"
 readme = "README.md"
 requires-python = ">=3.8, <3.13"
@@ -21,7 +21,7 @@ dependencies = [
     "fastapi>=0.115.0",
     "uvicorn>=0.31.0",
     "websockets>=13.1",
-    "anthropic>=0.36.2",
+    "pydantic>=2.9.2",
 ]
 
 [project.urls]

{wcgw-0.0.8 → wcgw-0.0.10}/src/wcgw/basic.py

@@ -16,13 +16,12 @@ from openai.types.chat import (
     ParsedChatCompletionMessage,
 )
 import rich
-import petname
+import petname  # type: ignore[import-untyped]
 from typer import Typer
 import uuid
 
-from wcgw.common import Config, text_from_editor
-
-from .common import Models
+from .common import Models, discard_input
+from .common import CostData, History
 from .openai_utils import get_input_cost, get_output_cost
 from .tools import ExecuteBash, ReadImage, ImageData
 
@@ -31,7 +30,6 @@ from .tools import (
     Confirmation,
     DoneFlag,
     Writefile,
-    get_is_waiting_user_input,
     get_tool_output,
     SHELL,
     start_shell,
@@ -40,14 +38,40 @@ from .tools import (
 import tiktoken
 
 from urllib import parse
+import subprocess
 import os
+import tempfile
 
 import toml
+from pydantic import BaseModel
 
 
 from dotenv import load_dotenv
 
-History = list[ChatCompletionMessageParam]
+
+class Config(BaseModel):
+    model: Models
+    secondary_model: Models
+    cost_limit: float
+    cost_file: dict[Models, CostData]
+    cost_unit: str = "$"
+
+
+def text_from_editor(console: rich.console.Console) -> str:
+    # First consume all the input till now
+    discard_input()
+    console.print("\n---------------------------------------\n# User message")
+    data = input()
+    if data:
+        return data
+    editor = os.environ.get("EDITOR", "vim")
+    with tempfile.NamedTemporaryFile(suffix=".tmp") as tf:
+        subprocess.run([editor, tf.name], check=True)
+        with open(tf.name, "r") as f:
+            data = f.read()
+            console.print(data)
+            return data
+
 
 def save_history(history: History, session_id: str) -> None:
     myid = str(history[1]["content"]).replace("/", "_").replace(" ", "_").lower()[:60]
@@ -67,29 +91,22 @@ def parse_user_message_special(msg: str) -> ChatCompletionUserMessageParam:
         if line.startswith("%"):
             args = line[1:].strip().split(" ")
             command = args[0]
-            assert command == 'image'
+            assert command == "image"
             image_path = args[1]
-            with open(image_path, 'rb') as f:
+            with open(image_path, "rb") as f:
                 image_bytes = f.read()
                 image_b64 = base64.b64encode(image_bytes).decode("utf-8")
                 image_type = mimetypes.guess_type(image_path)[0]
-                dataurl=f'data:{image_type};base64,{image_b64}'
-            parts.append({
-                'type': 'image_url',
-                'image_url': {
-                    'url': dataurl,
-                    'detail': 'auto'
-                }
-            })
+                dataurl = f"data:{image_type};base64,{image_b64}"
+            parts.append(
+                {"type": "image_url", "image_url": {"url": dataurl, "detail": "auto"}}
+            )
         else:
-            if len(parts) > 0 and parts[-1]['type'] == 'text':
-                parts[-1]['text'] += '\n' + line
+            if len(parts) > 0 and parts[-1]["type"] == "text":
+                parts[-1]["text"] += "\n" + line
             else:
-                parts.append({'type': 'text', 'text': line})
-    return {
-        'role': 'user',
-        'content': parts
-    }
+                parts.append({"type": "text", "text": line})
+    return {"role": "user", "content": parts}
 
 
 app = Typer(pretty_exceptions_show_locals=False)
@@ -121,7 +138,7 @@ def loop(
         if history[1]["role"] != "user":
             raise ValueError("Invalid history file, second message should be user")
         first_message = ""
-        waiting_for_assistant = history[-1]['role'] != 'assistant'
+        waiting_for_assistant = history[-1]["role"] != "assistant"
 
     my_dir = os.path.dirname(__file__)
     config_file = os.path.join(my_dir, "..", "..", "config.toml")
@@ -136,9 +153,6 @@ def loop(
     enc = tiktoken.encoding_for_model(
         config.model if not config.model.startswith("o1") else "gpt-4o"
     )
-    is_waiting_user_input = get_is_waiting_user_input(
-        config.model, config.cost_file[config.model]
-    )
 
     tools = [
         openai.pydantic_function_tool(
@@ -265,7 +279,7 @@ System information:
                     )
                     system_console.print(f"\nTotal cost: {config.cost_unit}{cost:.3f}")
                     output_toks += output_toks_
-                    
+
                     _histories.append(item)
                     for tool_call_id, toolcallargs in tool_call_args_by_id.items():
                         for toolindex, tool_args in toolcallargs.items():
@@ -275,7 +289,7 @@ System information:
                                     enc,
                                     limit - cost,
                                     loop,
-                                    is_waiting_user_input,
+                                    max_tokens=2048,
                                 )
                             except Exception as e:
                                 output_or_done = (
@@ -297,42 +311,49 @@ System information:
                                     f"\nTotal cost: {config.cost_unit}{cost:.3f}"
                                 )
                                 return output_or_done.task_output, cost
-                                
+
                             output = output_or_done
 
                             if isinstance(output, ImageData):
                                 randomId = petname.Generate(2, "-")
                                 if not image_histories:
-                                    image_histories.extend([
-                                        {
-                                            'role': 'assistant',
-                                            'content': f'Share images with ids: {randomId}'
-
-                                        },
-                                        {
-                                            'role': 'user',
-                                            'content': [{
-                                                'type': 'image_url',
-                                                'image_url': {
-                                                    'url': output.dataurl,
-                                                    'detail': 'auto'
-                                                }
-                                            }]
-                                        }]
+                                    image_histories.extend(
+                                        [
+                                            {
+                                                "role": "assistant",
+                                                "content": f"Share images with ids: {randomId}",
+                                            },
+                                            {
+                                                "role": "user",
+                                                "content": [
+                                                    {
+                                                        "type": "image_url",
+                                                        "image_url": {
+                                                            "url": output.dataurl,
+                                                            "detail": "auto",
+                                                        },
+                                                    }
+                                                ],
+                                            },
+                                        ]
                                     )
                                 else:
-                                    image_histories[0]['content'] += ', ' + randomId
-                                    image_histories[1]["content"].append({ # type: ignore
-                                        'type': 'image_url',
-                                        'image_url': {
-                                            'url': output.dataurl,
-                                            'detail': 'auto'
+                                    image_histories[0]["content"] += ", " + randomId
+                                    second_content = image_histories[1]["content"]
+                                    assert isinstance(second_content, list)
+                                    second_content.append(
+                                        {
+                                            "type": "image_url",
+                                            "image_url": {
+                                                "url": output.dataurl,
+                                                "detail": "auto",
+                                            },
                                         }
-                                    })
+                                    )
 
                                 item = {
                                     "role": "tool",
-                                    "content": f'Ask user for image id: {randomId}',
+                                    "content": f"Ask user for image id: {randomId}",
                                     "tool_call_id": tool_call_id + str(toolindex),
                                 }
                             else:

{wcgw-0.0.8 → wcgw-0.0.10}/src/wcgw/common.py

@@ -1,13 +1,9 @@
-import os
 import select
-import subprocess
 import sys
-import tempfile
 import termios
 import tty
 from typing import Literal
 from pydantic import BaseModel
-import rich
 
 
 class CostData(BaseModel):
@@ -15,6 +11,14 @@ class CostData(BaseModel):
     cost_per_1m_output_tokens: float
 
 
+from openai.types.chat import (
+    ChatCompletionMessageParam,
+    ChatCompletionAssistantMessageParam,
+    ChatCompletionMessage,
+    ParsedChatCompletionMessage,
+)
+
+History = list[ChatCompletionMessageParam]
 Models = Literal["gpt-4o-2024-08-06", "gpt-4o-mini"]
 
 
@@ -41,27 +45,3 @@ def discard_input() -> None:
     finally:
         # Restore old terminal settings
         termios.tcsetattr(fd, termios.TCSADRAIN, old_settings)
-
-
-class Config(BaseModel):
-    model: Models
-    secondary_model: Models
-    cost_limit: float
-    cost_file: dict[Models, CostData]
-    cost_unit: str = "$"
-
-
-def text_from_editor(console: rich.console.Console) -> str:
-    # First consume all the input till now
-    discard_input()
-    console.print("\n---------------------------------------\n# User message")
-    data = input()
-    if data:
-        return data
-    editor = os.environ.get("EDITOR", "vim")
-    with tempfile.NamedTemporaryFile(suffix=".tmp") as tf:
-        subprocess.run([editor, tf.name], check=True)
-        with open(tf.name, "r") as f:
-            data = f.read()
-            console.print(data)
-            return data