website·Indexed 31 days ago
https://starrailassistant.top/An automation tool for Honkai: Star Rail featuring SRA-cli for command-line task execution and SRA-server for RESTful API control. It supports a pluggable architecture for custom extensions, tasks, and commands, and includes built-in support for the Model Context Protocol (MCP) to allow AI assistants like Claude and Cursor to control the software.
help command to display information about a specific command. If the command name is omitted, the CLI will list all available commands.sra> help taskcmd2 library. When SRA starts, SRACore/__main__.py calls dynamic_import("tasks"), which scans all .py files in the tasks/ directory and imports them. Any class inheriting from cmd2.CommandSet[SRACli] is automatically instantiated and registered by SRACli (which has auto_load_commands=True), meaning no manual registration code is required as long as the file is placed in the tasks/ folder.SRA uses a pluggable architecture based on generic declarations and decorators. To create an extension, you must define a configuration model using Pydantic, inherit from BaseExtension[YourConfig], and register the class with the @extension decorator. Extensions are automatically discovered when placed in the extensions/ directory.
Key Components:
pydantic.BaseModel. Use Field to define defaults, descriptions (for the UI), and constraints (e.g., ge for minimum, le for maximum).@extension(name="...", description="...") to register the extension in the global ExtensionRegistry.run() method, which must return a boolean (True for success, False for failure).on_start(), on_completed(), and on_failed() for resource management and error recovery.from pydantic import BaseModel, Field
from SRACore.extension import BaseExtension, extension
# 1. Define Configuration
class MyConfig(BaseModel):
"""Extension configuration."""
target: str = Field(default="Star Rail", description="Target name")
count: int = Field(default=1, ge=1, le=100, description="Execution count")
verbose: bool = Field(default=False, description="Enable detailed logs")
# 2. Define Extension Class
@extension(name="My Extension", description="A custom extension example")
class MyExtension(BaseExtension[MyConfig]):
"""Custom extension implementation."""
def on_start(self) -> None:
print("Initializing resources...")
def run(self) -> bool:
for i in range(self.config.count):
print(f"[MyExtension] ({i + 1}/{self.config.count}) Target: {self.config.target}")
return True
def on_completed(self) -> None:
print("Cleaning up resources...")GitHub Repository: https://github.com/Shasnow/StarRailAssistant QQ Group 1: 994571792 QQ Group 2: 829477581
-1073741819), and references to Honkai: Star Rail and Wuthering Waves.Developer options are hidden by default and can be enabled by rapidly clicking the version number in the 'About' section. Once enabled, the following options become available:
main.py).tasks/ directory. The task must inherit from BaseTask and be marked with the @task decorator. SRA automatically scans this directory at startup and registers any matching classes. The execution index of a task is determined by the alphabetical order of the filenames in the tasks/ directory.from SRACore.task import BaseTask, task
from SRACore.util.logger import logger
@task
class MyCustomTask(BaseTask):
def run(self):
logger.info("This is my custom task!")
return True # Return True to indicate successful completion.srarc located in the current directory. The tool will automatically execute the commands listed in this file upon startup. Each command must be on a new line and will be executed sequentially until an exit command is encountered or the end of the file is reached.# Example .srarc file
run Default
exitDevelopers and users can contribute to StarRailAssistant by submitting bug reports, creating Pull Requests (PRs) for fixes or new features, improving documentation, or providing testing feedback across different devices and environments.
Recommended contribution workflow:
High-priority contributions include documentation corrections, new feature proposals, image recognition optimizations, and cross-platform/resolution testing feedback.