Configuration

Switchyard can be configured through its settings UI or by editing the config file directly.

Config File Location

• Standard: ~/.config/switchyard/config.toml
• Flatpak: ~/.var/app/io.github.alyraffauf.Switchyard/config/switchyard/config.toml

Example

prompt_on_click = true
favorite_browser = ''
check_default_browser = true

# Link redirections (domain type is default)
[[redirections]]
find = 'x.com'
replace = 'twitter.com'

[[redirections]]
find = 'reddit.com'
replace = 'old.reddit.com'

# Pattern redirection to remove query parameters
[[redirections]]
type = 'wildcard'
find = '?utm_*'
replace = ''

# Browser rule with a single condition
[[rules]]
name = 'Work GitHub'
browser = 'firefox.desktop'

[[rules.conditions]]
type = 'domain'
pattern = 'github.com'

# Multi-condition rule with AND logic
[[rules]]
name = 'Google Docs'
logic = 'all'  # all conditions must match
browser = 'google-chrome.desktop'

[[rules.conditions]]
type = 'domain'
pattern = 'docs.google.com'

[[rules.conditions]]
type = 'keyword'
pattern = 'edit'

# Multi-condition rule with OR logic
[[rules]]
name = 'Video Sites'
logic = 'any'  # any condition can match
browser = 'brave-browser.desktop'

[[rules.conditions]]
type = 'domain'
pattern = 'youtube.com'

[[rules.conditions]]
type = 'domain'
pattern = 'vimeo.com'

[[rules.conditions]]
type = 'domain'
pattern = 'twitch.tv'

# Rule with negated condition
[[rules]]
name = 'GitHub (not Gist)'
logic = 'all'
browser = 'firefox.desktop'

[[rules.conditions]]
type = 'glob'
pattern = '*.github.com'

[[rules.conditions]]
type = 'domain'
pattern = 'gist.github.com'
negate = true  # URL must NOT match this pattern

# Rule with always ask
[[rules]]
name = 'Shopping Sites'
always_ask = true

[[rules.conditions]]
type = 'keyword'
pattern = 'amazon'

Settings

• prompt_on_click: Show launcher when no rule matches (default: true).
• favorite_browser: Favorite browser that always appears first in launcher and is used as fallback when launcher is disabled.
• check_default_browser: Prompt to set Switchyard as system default browser on startup (default: true).

Browser Rules

Browser rules define how URLs are routed to browsers. Each rule has conditions that determine when it matches.

• name: Optional friendly name displayed in the UI.
• conditions: Array of conditions to match (see below).
• logic: How to combine conditions: all (AND) or any (OR). Default: all.
• browser: Desktop file ID of the target browser (e.g. firefox.desktop, com.google.Chrome.desktop).
• always_ask: If true, show browser launcher instead of auto-opening (default: false).

Conditions

Each condition specifies a pattern to match against the URL.

• type: Match type (see below).
• pattern: The pattern to match against.
• negate: If true, the condition is inverted (URL must NOT match). Default: false.

Condition Types

• domain: Exact domain match (e.g. github.com).
• keyword: URL contains text (e.g. youtube.com/watch).
• glob: Wildcard pattern with * (e.g. *.github.com).
• regex: Regular expression (e.g. ^https://.*\.example\.(com|org)).

Logic Modes

• all (AND): All conditions must match for the rule to apply.
• any (OR): Any single condition matching triggers the rule.

Use all for precise targeting (e.g., “docs.google.com AND contains ‘edit’”) and any for broad matching (e.g., “youtube.com OR vimeo.com OR twitch.tv”).

Link Redirections

Link redirections modify URLs before browser rules are evaluated.

• type: domain (default), wildcard, or regex.
• find: Pattern to match.
• replace: Text to replace with. Leave empty to remove matches.

Redirection Types

Domain redirections match the exact hostname, including the ports (if any). Use this to switch between sites:

[[redirections]]
find = 'reddit.com'
replace = 'old.reddit.com'

[[redirections]]
find = 'x.com'
replace = 'twitter.com'

Wildcard redirections match anywhere in the URL and support * wildcards. Use this to clean up links:

[[redirections]]
type = 'wildcard'
find = '?utm_*'
replace = ''

[[redirections]]
type = 'wildcard'
find = '&fbclid=*'
replace = ''

Regex redirections use regular expressions with capture group support ($1, $2, etc.) for complex transformations:

# Clean Amazon URLs - keep only the product ID
# amazon.com/dp/B0DZD91W4F/?tag=thewire06-20&linkCode=xm2&ascsubtag=... → amazon.com/dp/B0DZD91W4F
[[redirections]]
type = 'regex'
find = '(amazon\.[a-z.]+/dp/[A-Z0-9]+).*'
replace = '$1'

Redirections are applied in order. Domain and pattern matching is case-insensitive; regex matching is case-sensitive.