### Install mcpx-py as uv tool (Local) - Bash/Shell

Source: https://github.com/dylibso/mcpx-py/blob/main/README.md

This command installs the mcpx-py package as a command-line tool using uv's tool installation feature, installing from the current local directory.

```Bash
uv tool install .
```

--------------------------------

### Install mcpx-py as uv tool (PyPI) - Bash/Shell

Source: https://github.com/dylibso/mcpx-py/blob/main/README.md

This command installs the mcpx-py package as a command-line tool using uv's tool installation feature, pulling from PyPI.

```Bash
uv tool install mcpx-py
```

--------------------------------

### Start mcpx-client Chat - Bash/Shell

Source: https://github.com/dylibso/mcpx-py/blob/main/README.md

This command starts an interactive chat session with an LLM using the mcpx-client command-line tool.

```Bash
mcpx-client chat
```

--------------------------------

### Install mcpx-py as uv tool (Git) - Bash/Shell

Source: https://github.com/dylibso/mcpx-py/blob/main/README.md

This command installs the mcpx-py package as a command-line tool using uv's tool installation feature, pulling directly from the Git repository.

```Bash
uv tool install git+https://github.com/dylibso/mcpx-py
```

--------------------------------

### Install mcpx-py with uv - Bash/Shell

Source: https://github.com/dylibso/mcpx-py/blob/main/README.md

This command installs the mcpx-py Python library using the uv package manager. uv is a fast Python package installer and resolver.

```Bash
uv add mcpx-py
```

--------------------------------

### Get mcpx-client Help - Bash/Shell

Source: https://github.com/dylibso/mcpx-py/blob/main/README.md

This command displays the usage information and available options for the mcpx-client command-line tool.

```Bash
mcpx-client --help
```

--------------------------------

### Execute mcpx-client with uvx (PyPI) - Bash/Shell

Source: https://github.com/dylibso/mcpx-py/blob/main/README.md

This command executes the mcpx-client command-line tool directly from the mcpx-py package installed via PyPI using uvx, without requiring a global installation of the tool.

```Bash
uvx --from mcpx-py mcpx-client
```

--------------------------------

### Install mcpx-py with pip - Bash/Shell

Source: https://github.com/dylibso/mcpx-py/blob/main/README.md

This command installs the mcpx-py Python library using the standard pip package manager.

```Bash
pip install mcpx-py
```

--------------------------------

### Execute mcpx-client with uvx (Git) - Bash/Shell

Source: https://github.com/dylibso/mcpx-py/blob/main/README.md

This command executes the mcpx-client command-line tool directly from the mcpx-py package pulled from Git using uvx, without requiring a global installation of the tool.

```Bash
uvx --from git+https://github.com/dylibso/mcpx-py mcpx-client
```

--------------------------------

### Run Llamafile in JSON API Mode - Bash/Shell

Source: https://github.com/dylibso/mcpx-py/blob/main/README.md

This command executes a Llamafile model, starting it as a local server with a JSON API endpoint on a specified host and port.

```Bash
./your-model.llamafile --json-api --host 127.0.0.1 --port 8080
```

--------------------------------

### Use mcpx-py Chat with Sync Message (Structured Output) - Python

Source: https://github.com/dylibso/mcpx-py/blob/main/README.md

This Python code shows how to use the Chat class to get structured output by defining a Pydantic BaseModel and passing it to the `result_type` parameter. The response data will conform to the specified model.

```Python
from mcpx_py import Chat, BaseModel, Field
from typing import List

class Summary(BaseModel):
    """
    A summary of some longer text
    """
    source: str = Field("The source of the original_text")
    original_text: str = Field("The original text to be summarized")
    items: List[str] = Field("A list of summary points")

llm = Chat("claude-3-5-sonnet-latest", result_type=Summary)
response = llm.send_message_sync(
    "summarize the contents of example.com"
)
print(response.data)
```

--------------------------------

### List mcpx-client Tools - Bash/Shell

Source: https://github.com/dylibso/mcpx-py/blob/main/README.md

This command lists the available tools that can be called using the mcpx-client command-line tool.

```Bash
mcpx-client list
```

--------------------------------

### Use mcpx-py Chat with Sync Message (Basic) - Python

Source: https://github.com/dylibso/mcpx-py/blob/main/README.md

This Python code demonstrates how to initialize the Chat class with a specific LLM model and send a synchronous message. It prints the data received in the response.

```Python
from mcpx_py import Chat

llm = Chat("claude-3-5-sonnet-latest")

# Or OpenAI
# llm = Chat("gpt-4o")

# Or Ollama
# llm = Chat("ollama:qwen2.5")

# Or Gemini
# llm = Chat("gemini-2.0-flash")

response = llm.send_message_sync(
    "summarize the contents of example.com"
)
print(response.data)
```

--------------------------------

### Call mcpx-client Tool (eval-js) - Bash/Shell

Source: https://github.com/dylibso/mcpx-py/blob/main/README.md

This command demonstrates how to call a specific tool, `eval-js`, using the mcpx-client command-line tool, providing input as a JSON string.

```Bash
mcpx-client tool eval-js '{"code": "2+2"}'
```

--------------------------------

### Pull Ollama Model - Bash/Shell

Source: https://github.com/dylibso/mcpx-py/blob/main/README.md

This command uses the Ollama CLI to download a specific LLM model, such as `llama3.2`, for local use.

```Bash
ollama pull llama3.2
```

--------------------------------

### Make Llamafile Executable - Bash/Shell

Source: https://github.com/dylibso/mcpx-py/blob/main/README.md

This command changes the permissions of a downloaded Llamafile model to make it executable.

```Bash
chmod +x your-model.llamafile
```

--------------------------------

### Generate mcp.run Session ID (Write to file) - Bash/Shell

Source: https://github.com/dylibso/mcpx-py/blob/main/README.md

This command generates a new mcp.run session ID and automatically writes it to a configuration file that mcpx-py can use. It requires npm and the @dylibso/mcpx package.

```Bash
npx --yes -p @dylibso/mcpx gen-session --write
```

--------------------------------

### Generate mcp.run Session ID (Output to console) - Bash/Shell

Source: https://github.com/dylibso/mcpx-py/blob/main/README.md

This command generates a new mcp.run session ID and prints it to the console. This is useful if you need to store the session ID in an environment variable instead of a configuration file. It requires npm and the @dylibso/mcpx package.

```Bash
npx --yes -p @dylibso/mcpx gen-session
```

--------------------------------

### Set MCP_RUN_SESSION_ID Environment Variable - Bash/Shell

Source: https://github.com/dylibso/mcpx-py/blob/main/README.md

This command sets the MCP_RUN_SESSION_ID environment variable with the session ID obtained from the `gen-session` command run without the `--write` flag. This allows mcpx-py to authenticate with mcp.run.

```Bash
export MCP_RUN_SESSION_ID=kabA7w6qH58H7kKOQ5su4v3bX_CeFn4k.Y4l/s/9dQwkjv9r8t/xZFjsn2fkLzf+tkve89P1vKhQ
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.