Skip to content

Configuration

TrakFlow stores configuration in .trak_flow/config.json.

View Current Configuration

tf config list

Get a Value

tf config get actor

Set a Value

tf config set actor "username"

Configuration Options

General Settings

Key Type Default Description
actor string "" Current user/agent identifier
stealth boolean false Suppress status messages
no_push boolean false Disable auto-push to git

Import Settings

Key Type Default Description
import.orphan_handling string "allow" How to handle orphaned parent references

Validation Settings

Key Type Default Description
validation.strict boolean false Enable strict validation

ID Generation Settings

Key Type Default Description
id_generator.min_hash_length integer 8 Minimum length for generated IDs

MCP Server Settings

Key Type Default Description
mcp.port integer 3333 Default HTTP port for MCP server

Example Configuration

{
  "actor": "claude-agent",
  "stealth": false,
  "no_push": true,
  "import": {
    "orphan_handling": "allow"
  },
  "validation": {
    "strict": true
  },
  "id_generator": {
    "min_hash_length": 8
  },
  "mcp": {
    "port": 3333
  }
}

Nested Configuration

Access nested values with dot notation:

# Get nested value
tf config get import.orphan_handling

# Set nested value
tf config set mcp.port 4000

Environment Variables

Some settings can be overridden via environment variables:

Variable Description
TRAK_FLOW_ACTOR Override the actor setting
TRAK_FLOW_STEALTH Set to "1" for stealth mode

Reset to Defaults

Delete the config file to reset:

rm .trak_flow/config.json
tf init  # Recreates with defaults

Git Integration

The configuration file is git-tracked by default, allowing team members to share settings. The SQLite database is gitignored.

.trak_flow/trak_flow.db
.trak_flow/trak_flow.db-*

This is created automatically by tf init.