Sites and credentials

monloader does not parse any site itself; gallery-dl does. monloader ships curated profiles tuned for known sites, and falls back to a generic profile for anything else gallery-dl supports.

Curated profiles and the generic fallback

Curated profiles exist for 50-plus sites (boorus and manga/comic sites), listed on the Settings page in two tables. A profile knows the site’s booru family (which decides how its tags and ratings are read), the canonical post URL, what kind of login it needs, and a representative example URL for the test probe.
Everything else gallery-dl supports works through the generic fallback: per-category tags route by name where they match a monbooru category, ratings use tolerant rules, and the post URL is whatever gallery-dl reports. So a site works the day gallery-dl supports it; a profile only improves the result.

gallery-dl’s supported-sites list is the full menu. How each site’s metadata is translated is in metadata mapping.

The sites tables on the Settings page

Auth kinds

The login column on the Settings page shows what each site needs; a marker flags a site missing a credential it requires.

Auth kind	Shown in the login column as	Meaning	Examples
`none`	`none`	Reads without credentials	most boorus, most manga sites
`api_optional`	`api (opt)`	Works unauthenticated at low rates; a key raises rate limits	danbooru
`api_required`	`api key`	Refuses to read without a key	gelbooru, rule34
`cookies`	`cookies`	Needs a logged-in browser session exported to a cookies file	sankaku, exhentai

Per-site setup

A site’s edit dialog

For each site you use, the Settings page exposes:

Credential fields matching the family: an api key with a user id (gelbooru family), a username with an api key (danbooru and e621 families), or a cookies file path. Secrets are write-only - the page shows whether a value is set, never the value itself.
A target gallery dropdown, listing your monbooru galleries. Pushes from this site land there; leave it empty to use the default gallery.
Test runs a live probe against the profile’s example URL with your configured credentials, reporting ok, auth rejected, blocked, or failed in that row, with the underlying error on hover. A site still missing a required credential reads “needs cookies” or “needs api key”; a credential the site refuses reads “auth rejected”; a Cloudflare or captcha wall reads “blocked”. The row also shows when the site was last reached successfully; that note is kept in memory and resets on restart.
Reset drops the site’s credential block, reverting it to the profile defaults.

Saving a site rewrites the managed gallery-dl.json from the config; that file is never hand-edited.

Cookies files

For a cookies site, export your logged-in browser session as a cookies.txt file and place it under the cookies directory (/config/cookies by default). Set its path in the site’s Settings row, or directly in the TOML:

[[sites]]
name    = "sankaku"
cookies = "/config/cookies/sankaku.txt"
gallery = "default"

The equivalent TOML

The Settings rows write [[sites]] blocks in monloader.toml; you can edit them directly instead. name is the gallery-dl category:

[[sites]]
name         = "gelbooru"   # gallery-dl category
api_key      = ""
user_id      = ""
gallery      = "art"        # per-source target; empty = default gallery
lookup_order = 4            # lookup-chain position; 0 or absent = not queried

lookup_order decides whether (and when) the site is searched when finding tags by hash; the chain is edited from the Settings lookup section and explained in reverse lookup. The rest of the file is covered in configuration.