### Install yfinance and statsmodels

Source: https://github.com/microsoft/taskweaver/blob/main/README.md

Install the yfinance and statsmodels packages for financial forecasting examples.

```bash
pip install yfinance
pip install statsmodels
```

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

### Install and Run TaskWeaver Web UI

Source: https://context7.com/microsoft/taskweaver/llms.txt

Instructions for installing the Chainlit library and starting the TaskWeaver web-based UI for interactive conversations and visualizations.

```bash
# Install Chainlit
pip install chainlit

# Start the web UI
cd TaskWeaver/playground/UI
chainlit run app.py

# The UI will be available at http://localhost:8000
```

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

### Start vLLM Server with Guided Decoding

Source: https://github.com/microsoft/taskweaver/blob/main/website/blog/local_llm.md

Use this command to start the vLLM OpenAI compatible API server with guided decoding enabled via lm-format-enforcer. This is crucial for enforcing output formats from the LLM.

```shell
python -m vllm.entrypoints.openai.api_server --model meta-llama/Meta-Llama-3-8B-Instruct --guided-decoding-backend lm-format-enforcer
```

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

### Install Dependencies

Source: https://github.com/microsoft/taskweaver/blob/main/website/README.md

Run this command to install all required project dependencies.

```bash
$ yarn
```

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

### Install Dashscope Package

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/llms/qwen.md

Install the required dashscope library to enable QWen support.

```bash
pip install dashscope
```

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

### Install Groq package

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/llms/groq.md

Install the necessary Python package to enable Groq support.

```bash
pip install groq
```

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

### Install TaskWeaver

Source: https://github.com/microsoft/taskweaver/blob/main/README.md

Commands to clone the repository and install required Python dependencies.

```bash
# [optional to create conda environment]
# conda create -n taskweaver python=3.10
# conda activate taskweaver

# clone the repository
git clone https://github.com/microsoft/TaskWeaver.git
cd TaskWeaver
# install the requirements
pip install -r requirements.txt
```

```bash
pip install git+https://github.com/microsoft/TaskWeaver@<TAG>
```

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

### Launch the Web UI service

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/usage/webui.md

Navigate to the UI directory and start the Chainlit application.

```bash
# assume you are in the TaskWeaver folder
cd playground/UI/
# make sure you are in playground/UI/ folder
chainlit run app.py
```

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

### Start Local Development Server

Source: https://github.com/microsoft/taskweaver/blob/main/website/README.md

Launches a local development server with live reloading enabled.

```bash
$ yarn start
```

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

### Install Ollama and Serve LLMs Locally

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/llms/ollama.md

Use these bash commands to install Ollama on Linux/WSL2 and start serving LLM models. Ensure you have Ollama installed and a model served before configuring TaskWeaver.

```bash
## Install ollama on Linux & WSL2
curl https://ollama.ai/install.sh | sh
## Run the serving
ollama serve
```

```bash
ollama run llama2:13b
```

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

### Install Zhipu AI SDK

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/llms/glm.md

Install the required Python package to interface with Zhipu AI services.

```bash
pip install zhipuai
```

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

### Install Azure Identity Dependencies

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/llms/aoai.md

Install the required package for Entra authentication.

```bash
pip install azure-identity
```

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

### Install Langchain and Tabulate

Source: https://github.com/microsoft/taskweaver/blob/main/README.md

Install the Langchain and tabulate packages required for the SQL pull data plugin.

```bash
pip install langchain
pip install tabulate
```

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

### Example Usage of TaskWeaverApp

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/concepts/app.md

Demonstrates creating a TaskWeaverApp instance, getting a session, sending a message, and stopping the app. Ensure to call app.stop() before exiting.

```python
from taskweaver.app.app import TaskWeaverApp

app = TaskWeaverApp(app_dir="path/to/project/")
session = app.get_session()
round = session.send_message("Hello, how can I help you?")

print(round)
app.stop()

```

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

### Planner Conversation Example with Attachments

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/customization/example/example.md

A complex example demonstrating how the Planner communicates with the CodeInterpreter using plan-related attachments.

```yaml
enabled: True
rounds:
  - User_query: count the rows of /home/data.csv
    state: created
    post_list:
      - message: count the rows of /home/data.csv
        send_from: User
        send_to: Planner
        attachment_list:
      - message: Please load the data file /home/data.csv and count the rows of the loaded data
        send_from: Planner
        send_to: CodeInterpreter
        attachment_list:
        - type: init_plan
          content: |-
            1. load the data file
            2. count the rows of the loaded data <sequentially depends on 1>
            3. report the result to the User <interactively depends on 2>
        - type: plan
          content: |-
            1. instruct CodeInterpreter to load the data file and count the rows of the loaded data
            2. report the result to the User
        - type: current_plan_step
          content: 1. instruct CodeInterpreter to load the data file and count the rows of the loaded data
      - message: Load the data file /home/data.csv successfully and there are 100 rows in the data file
        send_from: CodeInterpreter
        send_to: Planner
        attachment_list:
      - message: The data file /home/data.csv is loaded and there are 100 rows in the data file
        send_from: Planner
        send_to: User
        attachment_list:
          - type: init_plan
            content: |-
              1. load the data file
              2. count the rows of the loaded data <sequentially depends on 1>
              3. report the result to the User <interactively depends on 2>
          - type: plan
            content: |-
              1. instruct CodeInterpreter to load the data file and count the rows of the loaded data
              2. report the result to the User
          - type: current_plan_step
            content: 2. report the result to the User
```

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

### Planner Reasoning Output Example

Source: https://github.com/microsoft/taskweaver/blob/main/website/blog/reasoning.md

Demonstrates the Planner's output format when incorporating reasoning steps before generating an initial plan.

```commandline
 Human   ▶  predict msft's stock price in the next 7 days
 ╭───< Planner >
 ├─► [plan_reasoning] The user wants to predict Microsoft's stock price for the next 7 days. This task involves time series forecasting, which requires historical stock price data and a forecasting model. However, the user hasn't provided any data or specified a model to use. I need to ask the user for more information.
 ├─► [init_plan]
 │   1. Request historical stock price data from the user
 │   2. Ask the user to specify a forecasting model or confirm if a default model should be used
 │   3. Load the historical stock price data <interactively depends on 1>
 │   4. Train the forecasting model using the historical data <interactively depends on 2 and 3>
```

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

### Install Langchain

Source: https://github.com/microsoft/taskweaver/blob/main/project/plugins/README.md

Install the Langchain library using pip. This is a prerequisite for plugins like paper_summary and sql_pull_data.

```bash
pip install langchain
```

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

### Plugin Execution Example

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/customization/plugin/how_to_develop_a_new_plugin.md

A sample conversation log demonstrating how the plugin is invoked and executed by the CodeInterpreter.

```bash
=========================================================
 _____         _     _       __
|_   _|_ _ ___| | _ | |     / /__  ____ __   _____  _____ 
  | |/ _` / __| |/ /| | /| / / _ \/ __ `/ | / / _ \/ ___/
  | | (_| \__ \   < | |/ |/ /  __/ /_/ /| |/ /  __/ /    
  |_|\__,_|___/_|\_\|__/|__/\___/\__,_/ |___/\___/_/     
=========================================================
TaskWeaver: I am TaskWeaver, an AI assistant. To get started, could you please enter your request?
Human: render ABC in ascii art
>>> [INIT_PLAN]
1. Render the text 'ABC' into ASCII art
>>> [PLAN]
1. Instruct CodeInterpreter to render the text 'ABC' into ASCII art
>>> [CURRENT_PLAN_STEP]
1. Instruct CodeInterpreter to render the text 'ABC' into ASCII art
>>> [SEND_TO]
CodeInterpreter
>>> [MESSAGE]
Please render the text 'ABC' into ASCII art
>>> [PLANNER->CODEINTERPRETER]
Please render the text 'ABC' into ASCII art
>>> [PYTHON]
# Call the plugin function to render the text 'ABC' into ASCII art
ascii_art_result = ascii_render("ABC")
ascii_art_result
>>> [VERIFICATION]
NONE
>>> [STATUS]
SUCCESS
>>> [RESULT]
The execution of the generated python code above has succeeded

The result of above Python code after execution is:
      ___           ___           ___     
     /\  \         /\  \         /\  \    
    /::\  \       /::\  \       /::\  \   
   /:/\:\  \     /:/\:\  \     /:/\:\  \  
  /::\~\:\  \   /::\~\:\__\   /:/  \:\  \ 
 /:/\:\ \:\__\ /:/\:\ \:|__| /:/__/ \:\__\
 \/__\:\/:/  / \:\~\:\/:/  / \:\  \  \/__/
      \::/  /   \:\ \::/  /   \:\  \      
      /:/  /     \:\/:/  /     \:\  \     
     /:/  /       \::/__/       \:\__\    
     \/__/         ~~            \/__/    

>>> [CODEINTERPRETER->PLANNER]
The following python code has been executed:
```

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

### Start TaskWeaver CLI

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/quickstart.md

Command to launch the TaskWeaver process and the expected CLI output.

```bash
# assume you are in the TaskWeaver folder cloned from the repository
python -m taskweaver -p ./project/ # -p is the path to the project directory
```

```bash
=========================================================
 _____         _     _       __
|_   _|_ _ ___| | _ | |     / /__  ____ __   _____  _____
  | |/ _` / __| |/ /| | /| / / _ \/ __ `/ | / / _ \/ ___/
  | | (_| \__ \   < | |/ |/ /  __/ /_/ /| |/ /  __/ /
  |_|\__,_|___/_|\_\|__/|__/\___/\__,_/ |___/\___/_/
=========================================================
TaskWeaver: I am TaskWeaver, an AI assistant. To get started, could you please enter your request?
Human: ___
```

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

### Initialize TaskWeaver Session

Source: https://context7.com/microsoft/taskweaver/llms.txt

Basic setup to initialize the TaskWeaver application and start a session with an event handler.

```python
app = TaskWeaverApp(app_dir="./project/")
session = app.get_session()

# Pass event handler to send_message
response = session.send_message(
    "Generate 10 random numbers",
    event_handler=DetailedEventHandler()
)
```

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

### Launch TaskWeaver

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/usage/cmd.md

Starts the TaskWeaver process using the specified project directory.

```bash
# assume you are in the TaskWeaver folder
python -m taskweaver -p ./project/ # -p is the path to the project directory
```

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

### Example Handcrafted Experience Content

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/customization/experience/handcrafted_experience.md

This is an example of the `experience_text` content, written in Markdown format, specifying rules for the agent.

```yaml
exp_id: 1
experience_text: |-
  - Say "world" if you hear "hello".
  - Say "peace" if you hear "love".
```

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

### Basic Conversation Example Structure

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/customization/example/example.md

A minimal example showing the structure of a conversation round between a User and a Planner.

```yaml
enabled: True
rounds:
   - User_query: ping
     state: created
     post_list:
     - message: ping
       send_from: User
       send_to: Planner
       attachment_list: []
     - message: pong
       send_from: Planner
       send_to: User
       attachment_list: []
```

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

### Start Tracing Infrastructure

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/advanced/telemetry.md

Use Docker Compose to launch the local OpenTelemetry collector, Jaeger, and Prometheus services.

```bash
cd /TaskWeaver/tracing
docker-compose up
```

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

### Start TaskWeaver CLI

Source: https://github.com/microsoft/taskweaver/blob/main/README.md

Command to launch the TaskWeaver process using a project directory.

```bash
# assume you are in the cloned TaskWeaver folder
python -m taskweaver -p ./project/
```

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

### Add Custom Packages to Dockerfile

Source: https://context7.com/microsoft/taskweaver/llms.txt

Example Dockerfile snippet demonstrating how to install additional Python packages for use within the TaskWeaver execution container.

```dockerfile
# docker/ces_container/Dockerfile - Add custom packages
FROM python:3.10-slim
# ... existing content ...

# Install additional packages for your plugins
RUN pip install --no-cache-dir --user sentence-transformers
RUN pip install --no-cache-dir --user scikit-learn
RUN pip install --no-cache-dir --user yfinance
```

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

### Install LiteLLM Proxy

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/llms/liteLLM.md

Install the LiteLLM package with proxy support. Ensure necessary environment variables for your chosen LLM are set, such as COHERE_API_KEY for Cohere.

```bash
pip install litellm[proxy]
```

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

### Task Decomposition and Tracking Example

Source: https://github.com/microsoft/taskweaver/blob/main/website/blog/reasoning.md

Displays the structure of a task plan, including initial plan steps, current plan, and the active step being executed.

```commandline
├─► [init_plan]
│   1. Request historical stock price data for MSFT from the user
│   2. Request the forecasting model or method to be used for prediction <interactively depends on 1>
│   3. Perform the prediction using the provided data and model <interactively depends on 2>
│   4. Report the prediction results to the user <interactively depends on 3>
├─► [plan]
│   1. Request historical stock price data for MSFT from the user
│   2. Request the forecasting model or method to be used for prediction
│   3. Perform the prediction using the provided data and model
│   4. Report the prediction results to the user
├─► [current_plan_step] 1. Request historical stock price data for MSFT from the user
```

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

### Update Code Interpreter Example Schema

Source: https://github.com/microsoft/taskweaver/blob/main/website/blog/local_llm.md

Examples showing the transition from deprecated attachment types to the new schema format.

```yaml
- type: python
  content: |-
    file_path = "/abc/def.txt"  

    with open(file_path, "r") as file:  
        file_contents = file.read()  
        print(file_contents)
```

```yaml
- type: reply_type
  content: python # or 'text' if the old type is 'text' or 'sample'
- type: reply_content
  content: |-
    file_path = "/abc/def.txt"  

    with open(file_path, "r") as file:  
        file_contents = file.read()  
        print(file_contents)
```

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

### Install Chainlit dependency

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/usage/webui.md

Install the required version of Chainlit to ensure compatibility with the Web UI.

```bash
pip install -U "chainlit<1.1.300"
```

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

### Verify Plugin Registration in Planner Prompt

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/FAQ.md

Example of the plugin registration section found in the planner prompt logs to confirm if a plugin is correctly loaded.

```markdown
CodeInterpreter has the following plugin functions and their required parameters need to be provided before the execution:
- anomaly_detection: anomaly_detection function identifies anomalies from an input DataFrame of time series. It will add a new column "Is_Anomaly", where each entry will be marked with "True" if the value is an anomaly or "False" otherwise. Arguments required: df: DataFrame, time_col_name: str, value_col_name: str
- ascii_render: This plugin renders the input text into ASCII art form. Arguments required: text: str
- klarna_search: Search and compare prices from thousands of online shops. Only available in the US. This plugin only takes user requests when searching for merchandise. If not clear, confirm with the user if they want to search for merchandise from Klarna. Arguments required: query: str
- tell_joke: Call this plugin to tell a joke.
```

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

### Run TaskWeaver Container

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/usage/docker.md

Execute this command to start a TaskWeaver container. Replace placeholders with your LLM API details.

```bash
docker run -it -e LLM_API_BASE=<API_BASE> \
  -e LLM_API_KEY=<API_KEY> \
  -e LLM_API_TYPE=<API_TYPE> \
  -e LLM_MODEL=<MODEL> \
  taskweavercontainers/taskweaver-all-in-one:latest
```

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

### Install agentops package

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/advanced/observability.md

Install the agentops package using pip. This is the first step to integrate AgentOps with your project.

```bash
pip install agentops
```

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

### Start TaskWeaver CLI

Source: https://context7.com/microsoft/taskweaver/llms.txt

Initiates the TaskWeaver CLI with a specified project directory. Provides an overview of available commands for interactive sessions.

```bash
# Start TaskWeaver CLI with default project directory
python -m taskweaver -p ./project/

# Example CLI session output:
# =========================================================
#  _____         _     _       __
# |_   _|_ _ ___| | _ | |     / /__  ____ __   _____  _____
#   | |/ _` / __| |/ /| | /| / / _ \/ __ `/ | / / _ \/ ___/
#   | | (_| \__ \   < | |/ |/ /  __/ /_/ /| |/ /  __/ /
#   |_|\__,_|___/_|\_\|__/|__/
# =========================================================
# TaskWeaver: I am TaskWeaver, an AI assistant. To get started, could you please enter your request?
# Human: ___
```

```bash
# Available CLI commands:
# /help           - Show help message
# /load <file>    - Load a file by absolute path
# /info           - Show current session information
# /reset          - Reset session and start fresh
# /clear          - Clear console output
# /save           - Save chat history for experience extraction
# /exit           - Exit the application

# Example: Load and analyze a CSV file
# Human: /load /home/user/data/sales.csv
# Human: Calculate the average sales per month and detect any anomalies
```

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

### SQL Query Generation Example

Source: https://github.com/microsoft/taskweaver/blob/main/website/blog/reasoning.md

SQL query generated by the Planner during the task execution.

```sql
SELECT EmployeeID, Department, ProjectCode, Role, Manager, Location
FROM EmployeeRecords
WHERE EmployeeID = 'E12345'
AND Department = 'Development'
AND ProjectCode = 'X987'
AND Role = 'Software Engineer'
AND Manager = 'J. Doe'
AND Location = 'Remote'
ORDER BY EmployeeID;
```

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

### TaskWeaver Recepta Reasoning Output

Source: https://github.com/microsoft/taskweaver/blob/main/website/blog/reasoning.md

Example output showing the interaction between the Planner and the Recepta role during a multi-step task.

```commandline
╭───< Planner >
├─► [plan_reasoning] The user has provided a set of instructions to follow step by step. The task involves extracting employee data from a summary, validating the role, and constructing an SQL query. I will follow these steps sequentially.
├─► [init_plan]
│   1. Extract the following information from the employee summary: `Employee ID`, `Department`, `Project Code`, `Role`, `Manager`, `Location`
│   2. Validate the role <interactively depends on 1>
│   3. Construct and print the SQL query using the extracted details <interactively depends on 1>
│   4. Check the execution result and report the result to the user <interactively depends on 1 and 3>
├─► [plan]
│   1. Extract information from the employee summary
│   2. Validate the role
│   3. Construct SQL query using the extracted details
│   4. Report the result to the user
├─► [current_plan_step] 1. Extract information from the employee summary
├─► [stop] InProcess
├──● Thought: I can read and comprehend the employee summary to extract the required information.
│   Result: the `Employee ID` is `E12345`, the `Department` is `Development`, the `Project Code` is `X987`, the `Role` is `Software Engineer`, the `Manager` is `J. Doe`, and the `Location` is `Remote`.
├─► [shared_memory_entry] Add the plan to the shared memory
╰──● sending message to Recepta
╭───< Recepta >
├──● Got it. Now, please go ahead!
╰──● sending message to Planner
╭───< Planner >
├─► [plan_reasoning] I have extracted the information from the employee summary. The next step is to verify if the role is one of the recognized roles.
├─► [init_plan]
│   1. Extract the following information from the employee summary: `Employee ID`, `Department`, `Project Code`, `Role`, `Manager`, `Location`
│   2. Validate the role <interactively depends on 1>
│   3. Construct and print the SQL query using the extracted details <interactively depends on 1>
│   4. Check the execution result and report the result to the user <interactively depends on 1 and 3>
├─► [plan]
│   1. Extract information from the employee summary
│   2. Validate the role
│   3. Construct SQL query using the extracted details
│   4. Report the result to the user
├─► [current_plan_step] 2. Validate the role
├─► [stop] InProcess
├──● Thought: I can verify if the role is one of the recognized roles with my pattern matching capability.
│   Thought: The recognized roles are `Software Engineer, Data Analyst, Project Manager, DevOps Specialist, QA Tester`.
│   Result: The role is `Software Engineer`, which is recognized.
├─► [shared_memory_entry] Add the plan to the shared memory
╰──● sending message to Recepta
╭───< Recepta >
├──● Got it. Now, please go ahead!
╰──● sending message to Planner
╭───< Planner >
├─► [plan_reasoning] I can prepare and print the SQL query using the extracted details to the User.
├─► [init_plan]
│   1. Extract the following information from the employee summary: `Employee ID`, `Department`, `Project Code`, `Role`, `Manager`, `Location`
│   2. Validate the role <interactively depends on 1>
│   3. Construct and print the SQL query using the extracted details <interactively depends on 1>
│   4. Check the execution result and report the result to the user <interactively depends on 1 and 3>
├─► [plan]
│   1. Extract information from the employee summary
│   2. Validate the role
│   3. Construct SQL query using the extracted details
│   4. Report the result to the user
├─► [current_plan_step] 3. Construct and print the SQL query using the extracted details and report the result to the user
├─► [stop] Completed
├──● The role is `Software Engineer`, which is recognized.
│
│   Here is the SQL query using the extracted details:
│   ```sql
│   SELECT EmployeeID, Department, ProjectCode, Role, Manager, Location
│   FROM EmployeeRecords
│   WHERE EmployeeID = 'E12345'
│   AND Department = 'Development'
│   AND ProjectCode = 'X987'
│   AND Role = 'Software Engineer'
│   AND Manager = 'J. Doe'
│   AND Location = 'Remote'
│   ORDER BY EmployeeID;
│   ```
├─► [shared_memory_entry] Add the plan to the shared memory
╰──● sending message to User
 TaskWeaver ▶  The role is `Software Engineer`, which is recognized.
```

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

### Run TaskWeaver with Project Directory Mount

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/usage/docker.md

Mount your local project directory into the container to use custom configurations, plugins, and examples.

```bash
docker run -it -e LLM_API_BASE=<API_BASE> \
  -e LLM_API_KEY=<API_KEY> \
  -e LLM_API_TYPE=<API_TYPE> \
  -e LLM_MODEL=<MODEL> \
#  -e TASKWEAVER_UID=$(id -u) \ # uncomment if your host OS is not Windows
#  -e TASKWEAVER_GID=$(id -g) \ # uncomment if your host OS is not Windows
  --mount type=bind,source=<your_local_project_dir>,target=/app/TaskWeaver/project/ \
  taskweavercontainers/taskweaver-all-in-one:latest
```

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

### Run LiteLLM Proxy Server

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/llms/liteLLM.md

Start the LiteLLM proxy server. Use the `--drop_params` argument for API compatibility. The server will be accessible at http://0.0.0.0:8000.

```bash
litellm --model MODEL_NAME --drop_params
```

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

### No-Planner Mode Execution Example

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/configurations/configurations_in_detail.md

Demonstrates a direct interaction with the CodeInterpreter in no-planner mode, showing the input request and the resulting Python code execution output.

```bash
 =========================================================
     _____         _     _       __
    |_   _|_ _ ___| | _ | |     / /__  ____ __   _____  _____
      | |/ _` / __| |/ /| | /| / / _ \/ __ `/ | / / _ \/ ___/
      | | (_| \__ \   < | |/ |/ /  __/ /_/ /| |/ /  __/ /
      |_|\__,_|___/_|\_\|__/|__/\___/\__,_/ |___/\___/_/
    =========================================================
    TaskWeaver: I am TaskWeaver, an AI assistant. To get started, could you please enter your request?
    Human: generate 10 random numbers
    >>> [PYTHON]Starting... 
    import numpy as np
    random_numbers = np.random.rand(10)
    random_numbers
    >>> [VERIFICATION]
    NONE
    >>> [STATUS]Starting...         
    SUCCESS
    >>> [RESULT]
    The execution of the generated python code above has succeeded
    
    The result of above Python code after execution is:
    array([0.09918602, 0.68732778, 0.44413814, 0.4756623 , 0.48302334,
           0.8286594 , 0.80994359, 0.35677263, 0.45719317, 0.68240194])
    >>> [CODEINTERPRETER->PLANNER]
    The following python code has been executed:
    ```python
    import numpy as np
    random_numbers = np.random.rand(10)
    random_numbers
    ```
    
    The execution of the generated python code above has succeeded
    
    The result of above Python code after execution is:
    array([0.09918602, 0.68732778, 0.44413814, 0.4756623 , 0.48302334,
           0.8286594 , 0.80994359, 0.35677263, 0.45719317, 0.68240194])
    TaskWeaver: The following python code has been executed:
    ```python
    import numpy as np
    random_numbers = np.random.rand(10)
    random_numbers
    ```
  
    The execution of the generated python code above has succeeded
    
    The result of above Python code after execution is:
    array([0.09918602, 0.68732778, 0.44413814, 0.4756623 , 0.48302334,
           0.8286594 , 0.80994359, 0.35677263, 0.45719317, 0.68240194])
```

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

### Docker Compose Infrastructure Setup

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/advanced/telemetry.md

Defines the services for the OpenTelemetry collector, Jaeger, and Prometheus.

```yaml
version: '3'
services:
  optl-collector:
    image: otel/opentelemetry-collector:0.96.0
    command: ["--config=/etc/collector-config.yaml"]
    volumes:
      - ./collector-config.yaml:/etc/collector-config.yaml
    ports:
      - "4317:4317" # Expose the gRPC receiver port for the collector
    depends_on:
      - jaeger

  jaeger:
    image: jaegertracing/all-in-one:1.54
    ports:
      - "16686:16686" # Jaeger UI

  prometheus:
    image: prom/prometheus:latest
    ports:
      - "9090:9090" # Prometheus UI
    volumes:
      - ./prometheus-config.yml:/etc/prometheus/prometheus.yml
    command: ["--config.file=/etc/prometheus/prometheus.yml"]
    depends_on:
      - optl-collector
```

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

### Initialize TaskWeaverApp

Source: https://context7.com/microsoft/taskweaver/llms.txt

Initialize TaskWeaverApp with a project directory or inline configuration. Get a session, send a message, and stop the app when done.

```python
from taskweaver.app.app import TaskWeaverApp

# Initialize TaskWeaver with a project directory
app_dir = "./project/"  # Directory containing taskweaver_config.json
app = TaskWeaverApp(app_dir=app_dir)

# Alternative: Initialize with inline configuration
app = TaskWeaverApp(
    app_dir="./project/",
    config={
        "llm.api_key": "your-api-key",
        "llm.model": "gpt-4",
        "llm.api_type": "openai"
    }
)

# Get a new session for conversation
session = app.get_session()

# Send a message and get response
response_round = session.send_message("What can you do?")

# Access the response
print(response_round.to_dict())
# Output: {
#     "id": "round-20231201-043134-218a2681",
#     "user_query": "What can you do?",
#     "state": "finished",
#     "post_list": [...] 
# }

# Stop the app when done (must be called before exit)
session.stop()
app.stop()
```

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

### Verify Plugin Availability via LLM Response

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/FAQ.md

Example of the expected LLM response when querying TaskWeaver for available capabilities.

```markdown
I can assist you with various tasks, including:

- Detecting anomalies in time series data.
- Rendering text into ASCII art.
- Searching and comparing prices from thousands of online shops (US only).
- Telling a joke.

If you have a specific task in mind, please let me know, and I'll do my best to assist you.
```

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

### Load Local File in TaskWeaver CLI

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/usage/docker.md

Example of loading a local file (e.g., example.csv) into TaskWeaver using the `/load` command within the CLI.

```bash
 TaskWeaver ▶  I am TaskWeaver, an AI assistant. To get started, could you please enter your request?
    Human   ▶  /load /app/TaskWeaver/local/example.csv
    Human   ▶  display the column names of the loaded file
```

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

### Start TaskWeaver with Auto Plugin Selection

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/advanced/plugin_selection.md

Initiates TaskWeaver with the auto plugin selection feature enabled. This is the initial step to observe the mechanism in action.

```bash
=========================================================
 _____         _     _       __
|_   _|_ _ ___| | _ | |     / /
  | |/ _` / __| |/ /| | /| / / _ 
  | | (_| 
  |_|
=========================================================
TaskWeaver: I am TaskWeaver, an AI assistant. To get started, could you please enter your request?
Human: 
```

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

### Identify YAML loading errors

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/FAQ.md

Example of the error message displayed in logs when a plugin's YAML configuration file fails to load.

```bash
failed to loading component from <name>.yaml, skipping: Yaml loading failed due to: <reason>
```

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

### Example Plugin Usage in TaskWeaver

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/customization/plugin/plugin_intro.md

This simplified code demonstrates how TaskWeaver might call two plugins: `sql_pull_data` to fetch data and `anomaly_detection` to process it.

```python
df, data_description = sql_pull_data(query="pull data from time_series table")  
anomaly_df, anomaly_description = anomaly_detection(df, time_col_name="ts", value_col_name="val") 
```

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

### Test LLM API Integration

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/llms/customized_llm_api.md

After configuring `taskweaver_config.json`, navigate to the `./scripts` directory and run the test command to verify your LLM API setup.

```bash
cd ./scripts
```

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

### Call a plugin in plugin-only mode

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/advanced/plugin_only.md

Example of the restricted code format that only invokes a plugin.

```python
r1=ascii_render(**{"text":"Hello"})
r1
```

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

### Import agentops package

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/advanced/observability.md

Import the agentops package into your Python project to start using its functionalities.

```python
import agentops
```

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

### Configure Planner Experience Directory

Source: https://github.com/microsoft/taskweaver/blob/main/website/blog/experience.md

Use this JSON configuration to enable and set a custom experience directory for the Planner role. Ensure the `planner.use_experience` is set to true.

```json
{
  "planner.use_experience": true,
  "planner.experience_dir": "planner_exp_dir"
}
```

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

### Get Session

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/concepts/app.md

Retrieves an existing session by its ID or creates a new one if no ID is provided.

```APIDOC
## Get Session

### Description
Retrieves a session associated with the TaskWeaver app. If a `session_id` is provided, it returns the corresponding session. Otherwise, it creates and returns a new session.

### Method
get_session

### Endpoint
N/A (This is a method of the TaskWeaverApp class)

### Parameters
#### Path Parameters
None

#### Query Parameters
None

#### Request Body
None

### Parameters
- **session_id** (Optional[str]) - The ID of the session to retrieve. If `None`, a new session is created.

### Request Example
```python
# Assuming 'app' is an instance of TaskWeaverApp
session = app.get_session(session_id="existing_session_123")
# Or to create a new session:
session = app.get_session()
```

### Response
#### Success Response (200)
- **Session** (Session) - The retrieved or newly created session object.
```

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

### TaskWeaver Configuration with Custom Role

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/concepts/role.md

Example of the taskweaver_config.json file enabling the 'echo' role alongside default roles.

```json
{
    "session.roles": ["planner", "echo", "code_interpreter"]
}

```

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

### Prepare DA-Bench Test Cases

Source: https://github.com/microsoft/taskweaver/blob/main/auto_eval/dabench_scripts/README.md

Use this script to generate test cases for the DA-Bench benchmark. Ensure you provide the correct paths for the questions file, label file, data folder, and the desired output directory.

```bash
cd auto_eval/dabench_scripts 
python prepare_cases.py <path_to_the_jsonl_questions_file> <path_to_thejsonl_label_file> <path_to_the_data_folder> <output_directory>
```

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

### Example DA-Bench Test Case Prompt

Source: https://github.com/microsoft/taskweaver/blob/main/auto_eval/dabench_scripts/README.md

This markdown illustrates the structure of a test case prompt, including the task, question, constraints, and the expected output format. Adhere to this format when creating or interpreting test cases.

```markdown
# Task
Load the file test_ave.csv and answer the following questions.

# Question
Calculate the mean fare paid by the passengers.

   
# Constraints
Calculate the mean fare using Python's built-in statistics
module or appropriate statistical method in pandas. Rounding off the answer to
two decimal places.

# Format
@mean_fare[mean_fare_value] where "mean_fare_value" is a floating-point number rounded to two decimal places.
```

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

### Initialize TaskWeaverApp

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/concepts/app.md

Initialize the TaskWeaver app with optional project directory, local URI preference, and configuration settings.

```python
class TaskWeaverApp(object):
    def __init__(
        self,
        app_dir: Optional[str] = None,
        use_local_uri: Optional[bool] = None,
        config: Optional[Dict[str, Any]] = None,
        **kwargs: Any,
    ) -> None:
        """
        Initialize the TaskWeaver app.
        :param app_dir: The project directory.
        :param use_local_uri: Whether to use local URI for artifacts.
        :param config: The configuration.
        :param kwargs: The additional arguments.
        """

```

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

### Get Session from TaskWeaverApp

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/concepts/app.md

Retrieve a session from the TaskWeaver app. A new session is created if no session ID is provided.

```python
def get_session(
    self,
    session_id: Optional[str] = None,
) -> Session:
    """
    Get the session. Return a new session if the session ID is not provided.
    :param session_id: The session ID.
    :return: The session.
    """

```

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

### Example Docker network inspection output

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/code_execution.md

Shows the expected JSON output when inspecting a Docker network to identify its subnet.

```json
"Config": [
    {
        "Subnet": "172.19.0.0/16",
        "Gateway": "172.19.0.1"
    }
]
```

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

### Run TaskWeaver with Local Directory Mount

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/usage/docker.md

Mount a local directory to access files within the container at `/app/TaskWeaver/local/`. Use the `/load` command to access them.

```bash
docker run -it -e LLM_API_BASE=<API_BASE> \
  -e LLM_API_KEY=<API_KEY> \
  -e LLM_API_TYPE=<API_TYPE> \
  -e LLM_MODEL=<MODEL> \
#  -e TASKWEAVER_UID=$(id -u) \ # uncomment if your host OS is not Windows
#  -e TASKWEAVER_GID=$(id -g) \ # uncomment if your host OS is not Windows
  --mount type=bind,source=<your_local_dir>,target=/app/TaskWeaver/local/ \
  taskweavercontainers/taskweaver-all-in-one:latest
```

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

### Initialize TaskWeaver and Send Messages

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/usage/library.md

Use TaskWeaverApp to create a session and send user queries. The app_dir must point to the directory containing taskweaver_config.json.

```python
from taskweaver.app.app import TaskWeaverApp

# This is the folder that contains the taskweaver_config.json file and not the repo root. Defaults to "./project/"
app_dir = "./project/"
app = TaskWeaverApp(app_dir=app_dir)
session = app.get_session()

user_query = "hello, what can you do?"
response_round = session.send_message(user_query)
print(response_round.to_dict())
```

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

### Build Document Index via CLI

Source: https://github.com/microsoft/taskweaver/blob/main/taskweaver/ext_role/document_retriever/README.md

Execute the document indexer script to process documentation files into a searchable index.

```bash
cd TaskWeaver
python script/document_indexer.py \
  --doc_paths website/docs website/blog \
  --output_path project/sample_data/knowledge_base \
  --extensions md
```

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

### Round Object Structure

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/usage/library.md

Example of the JSON structure returned by a session round, containing the user query, state, and a list of posts.

```json
{
    "id": "round-20231201-043134-218a2681",
    "user_query": "hello, what can you do?",
    "state": "finished",
    "post_list": [
        {
            "id": "post-20231201-043134-10eedcca",
            "message": "hello, what can you do?",
            "send_from": "User",
            "send_to": "Planner",
            "attachment_list": []
        },
        {
            "id": "post-20231201-043141-86a2aaff",
            "message": "I can help you with various tasks, such as counting rows in a data file, detecting anomalies in a dataset, searching for products on Klarna, summarizing research papers, and pulling data from a SQL database. Please provide more information about the task you want to accomplish, and I'll guide you through the process.",
            "send_from": "Planner",
            "send_to": "User",
            "attachment_list": [
                {
                    "id": "atta-20231201-043141-6bc4da86",
                    "type": "init_plan",
                    "content": "1. list the available functions"
                },
                {
                    "id": "atta-20231201-043141-6f29f6c9",
                    "type": "plan",
                    "content": "1. list the available functions"
                },
                {
                    "id": "atta-20231201-043141-76186c7a",
                    "type": "current_plan_step",
                    "content": "1. list the available functions"
                }
            ]
        }
    ]
}
```

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

### Example of Potentially Dangerous Generated Code

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/advanced/verification.md

This snippet demonstrates code that reads and executes an external file, which is a target for verification rules.

```python
import os

file = open("abc.py", "r")
code = file.read()
exec(code)
```

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

### Build Static Content

Source: https://github.com/microsoft/taskweaver/blob/main/website/README.md

Compiles the website into static files located in the build directory.

```bash
$ yarn build
```

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

### TaskWeaver Log Output - Plugin Initialization

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/advanced/plugin_selection.md

Shows log entries indicating successful initialization of the Planner and the generation of plugin embeddings, confirming the auto plugin selector is ready.

```log
2023-12-18 14:23:44,197 - INFO - Planner initialized successfully
2023-12-18 14:24:10,488 - INFO - Plugin embeddings generated
2023-12-18 14:24:10,490 - INFO - CodeInterpreter initialized successfully.
2023-12-18 14:24:10,490 - INFO - Session 20231218-062343-c18494b1 is initialized
```

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

### Define an Anomaly Detection Plugin Schema

Source: https://github.com/microsoft/taskweaver/blob/main/website/docs/customization/plugin/plugin_intro.md

Example of a complete YAML schema for an anomaly detection plugin, including parameters and return values.

```yaml
name: anomaly_detection
enabled: true
plugin_only: false
required: false
description: >-
  anomaly_detection function identifies anomalies from an input DataFrame of
  time series. It will add a new column "Is_Anomaly", where each entry will be marked with "True" if the value is an anomaly or "False" otherwise.

parameters:
  - name: df
    type: DataFrame
    required: true
    description: >-
      the input data from which we can identify the anomalies with the 3-sigma
      algorithm.
  - name: time_col_name
    type: str
    required: true
    description: name of the column that contains the datetime
  - name: value_col_name
    type: str
    required: true
    description: name of the column that contains the numeric values.

returns:
  - name: df
    type: DataFrame
    description: >-
      This DataFrame extends the input DataFrame with a newly-added column
      "Is_Anomaly" containing the anomaly detection result.
  - name: description
    type: str
    description: This is a string describing the anomaly detection results.
```