Configuration
Below is a list of available configuration options.
Initial Setup
The default configuration file is located at ~/.dasshh/config.yaml
.
Below command helps you initialize the configuration file.
Configuration Options
Dasshh Configuration
Option | Description |
---|---|
skip_summarization |
Skip summarization of tool call results |
system_prompt |
The system prompt for the assistant |
tool_directories |
The directories to search for tools |
Model Configuration
Option | Description |
---|---|
model.name |
The model provider and model name |
model.api_key |
Your API key for the chosen provider |
model.api_base |
The base URL for the chosen provider |
model.api_version |
The version of the API to use |
model.temperature |
The temperature for the model |
model.top_p |
The top-p value for the model |
model.max_tokens |
The maximum number of tokens to generate |
model.max_completion_tokens |
The maximum number of tokens to generate for the completion |
Example Configuration File
The configuration file is a YAML file with the following structure.
# Dasshh configuration
dasshh:
skip_summarization: false
system_prompt: |
You are a helpful assistant that can help with tasks on the system.
Your goal is to save user's time by performing tasks on their behalf.
tool_directories:
- /Users/viiyer/repos/dasshh/dasshh/apps
# Model configuration
model:
name: gemini/gemini-2.0-flash
api_base:
api_key: <your-google-AI-studio-api-key>
api_version:
temperature: 1.0
top_p: 1.0
max_tokens: 1000
max_completion_tokens: 1000
Supported Models
Dasshh supports a variety of models through LiteLLM.