config_dev.yaml

CONTROL_BACKEND: "uia"  # The backend for control action, currently we support uia and win32
MAX_STEP: 100  # The max step limit for completing the user request
SLEEP_TIME: 5  # The sleep time between each step to wait for the window to be ready
RECTANGLE_TIME: 1

SAFE_GUARD: True  # Whether to use the safe guard to prevent the model from doing sensitve operations.
CONTROL_LIST: ["Button", "Edit", "TabItem", "Document", "ListItem", "MenuItem", "ScrollBar", "TreeItem", "Hyperlink", "ComboBox", "RadioButton", "DataItem"] 
# The list of widgets that allowed to be selected, in uia backend, it will be used for filter the control_type, while in win32 backend, it will be used for filter the class_name.
HISTORY_KEYS: ["Step", "Thought", "ControlText", "Subtask", "Action", "Comment", "Results", "UserConfirm"]  # The keys of the action history for the next step.
ANNOTATION_COLORS: {
        "Button": "#FFF68F",
        "Edit": "#A5F0B5",
        "TabItem": "#A5E7F0",
        "Document": "#FFD18A",
        "ListItem": "#D9C3FE",
        "MenuItem": "#E7FEC3",
        "ScrollBar": "#FEC3F8",
        "TreeItem": "#D6D6D6",
        "Hyperlink": "#91FFEB",
        "ComboBox": "#D8B6D4"
    }

PRINT_LOG: False  # Whether to print the log  
CONCAT_SCREENSHOT: False  # Whether to concat the screenshot for the control item
LOG_LEVEL: "DEBUG"  # The log level
INCLUDE_LAST_SCREENSHOT: True  # Whether to include the last screenshot in the observation
REQUEST_TIMEOUT: 250  # The call timeout for the GPT-V model

HOSTAGENT_PROMPT: "ufo/prompts/share/base/host_agent.yaml"  # The prompt for the app selection
# Due to the limitation of input size, lite version of the prompt help users have a taste. And the path is "ufo/prompts/share/lite/host_agent.yaml"
APPAGENT_PROMPT: "ufo/prompts/share/base/app_agent.yaml"  # The prompt for the action selection
# Lite version: "ufo/prompts/share/lite/app_agent.yaml"
FOLLOWERAHENT_PROMPT: "ufo/prompts/share/base/app_agent.yaml"  # The prompt for the follower agent

EVALUATION_PROMPT: "ufo/prompts/evaluation/evaluate.yaml"  # The prompt for the evaluation

HOSTAGENT_EXAMPLE_PROMPT: "ufo/prompts/examples/{mode}/host_agent_example.yaml"  # The prompt for the app selection
# Lite version: "ufo/prompts/examples/lite/{mode}/host_agent_example.yaml"
APPAGENT_EXAMPLE_PROMPT: "ufo/prompts/examples/{mode}/app_agent_example.yaml"  # The prompt for the action selection
# Lite version: "ufo/prompts/examples/lite/{mode}/app_agent_example.yaml"

## For experience learning
EXPERIENCE_PROMPT: "ufo/prompts/experience/experience_summary.yaml"
EXPERIENCE_SAVED_PATH: "vectordb/experience/"

## For user demonstration learning
DEMONSTRATION_PROMPT: "ufo/prompts/demonstration/demonstration_summary.yaml"
DEMONSTRATION_SAVED_PATH: "vectordb/demonstration/"

API_PROMPT: "ufo/prompts/share/base/api.yaml"  # The prompt for the API
CLICK_API: "click_input" # The click API
INPUT_TEXT_API: "type_keys" # The input text API. Can be "type_keys" or "set_text"
INPUT_TEXT_ENTER: False # whether to press enter after typing the text


## APIs related
USE_APIS: True  # Whether to use the API
WORD_API_PROMPT: "ufo/prompts/apps/word/api.yaml"  # The prompt for the word API
EXCEL_API_PROMPT: "ufo/prompts/apps/excel/api.yaml"  # The prompt for the word API

# For control filtering
#'TEXT' for only rich text filter, 'SEMANTIC' for only semantic similarity match, 'ICON' for only icon match
CONTROL_FILTER_TYPE: []  # The list of control filter type, support 'TEXT', 'SEMANTIC', 'ICON'
CONTROL_FILTER_TOP_K_PLAN: 2  # The control filter effect on top k plans from UFO, default is 2
CONTROL_FILTER_TOP_K_SEMANTIC: 15  # The control filter top k for semantic similarity
CONTROL_FILTER_TOP_K_ICON: 15  # The control filter top k for icon similarity
CONTROL_FILTER_MODEL_SEMANTIC_NAME: "all-MiniLM-L6-v2"  # The control filter model name of semantic similarity
CONTROL_FILTER_MODEL_ICON_NAME: "clip-ViT-B-32"  # The control filter model name of icon similarity

ALLOW_OPENAPP: FALSE  # Whether to allow the open app action
LOG_XML: False  # Whether to log the xml file for the at every step.
SCREENSHOT_TO_MEMORY: True  # Whether to allow the screenshot to memory for the agent's decision making.


# For customizations
USE_CUSTOMIZATION: True  # Whether to use the customization
QA_PAIR_FILE: "customization/historical_qa.txt"  # The path for the historical QA
QA_PAIR_NUM: 20  # The number of QA pairs for the customization

# For the evaluation
EVA_SESSION: True  # Whether to include the session in the evaluation
EVA_ROUND: FALSE
EVA_ALL_SCREENSHOTS: True  # Whether to include all the screenshots in the evaluation