Skip to content
MCP server

danbooru-tag-mcp

MCP server that looks up Danbooru tags, aliases and post stats for AI image prompts

Actively maintainedStatus and history

The repository was last updated 7 days ago and is published under MIT.

Pricing
Open source
Platforms
Windows, CLI
GitHub stars
1
Last commit
7 days ago
Language
Go
License
MIT

Similar MCP servers

About danbooru-tag-mcp

danbooru-tag-mcp is an open-source MCP server for design and media. It is available on Windows and CLI.

Rather than having a model recall Danbooru tag names from memory, this server lets an AI client query the live booru database for exact names, aliases, categories, post counts, co-occurrence statistics and wiki rules before it writes a prompt. The README says it also checks each name against pinned WD14 tagger vocabularies, so captions using older or renamed tags can be traced to their current form.

The server exposes eight tools, including a batch lookup that resolves up to 20 names in one call and a tag profile that samples posts to show what a tag typically renders alongside. Responses are cached on disk for stale-tolerant fallback during outages, and requests are spaced to keep anonymous access workable. Tags are returned as plain strings that can be pasted into ComfyUI, Stable Diffusion or Illustrious prompt fields.

It is aimed at people running local image generation on Windows who want their prompts built from tags that actually exist on Danbooru.

Tools this server provides (8)

Taken from the project documentation; the running server may differ.

search_tags
Search tags by keyword, ordered by post count
get_tag_info
Exact info for one tag, with a WD14 verdict on the name
get_tags_info
Exact info for up to 20 names in one request
get_related_tags
Tags that co-occur with a given tag
get_tag_alias
Resolve an alias or misspelling to its canonical tag
get_tag_wiki
A tag's wiki page as structured fields
search_posts
Posts matching a tag combination, with their tag lists
get_tag_profile
What a tag actually renders as, from a post sample

What it does

Exact tag lookup
Returns canonical names, categories and post counts so prompts use tags that exist.
Batch name resolution
Checks up to 20 tag names in one request, splitting found and missing results.
Tag distribution profiles
Shows co-occurring tags and high-lift combinations sampled from a tag's posts.
WD14 tagger verdicts
Classifies names as live, tagger-era or unknown against pinned WD14 vocabularies.
Structured wiki pages
Breaks Danbooru wiki pages into summary, rules, see-also and example posts.
Offline-tolerant cache
Serves the last cached answer when Danbooru is unreachable, marked as stale.

Good for

  • Resolving the correct Danbooru name for a visual trait before writing an image prompt
  • Checking whether tags from a LoRA caption still exist or were renamed
  • Finding tags that commonly appear together with a given tag
  • Reading wiki usage rules to avoid conflicting tags in a prompt
  • Sampling recent posts to see how a tag combination is typically tagged

Pricing

Open source

Free and MIT licensed; no paid tier mentioned.

danbooru-tag-mcp FAQ

Do I need a Danbooru account to use it?

No. The README says anonymous access works. Setting DANBOORU_LOGIN and DANBOORU_API_KEY is optional and gives higher rate limits and content access.

Which operating systems are supported?

The README lists Windows 10 and 11 on x64 or ARM64 only.

How are tags returned to the client?

As plain strings with no model coupling, so they can be pasted directly into ComfyUI, Stable Diffusion or Illustrious prompt fields.

Are post search and profile results cached?

No. The README says post search and tag profiles are never cached.

Is danbooru-tag-mcp free?

Yes. danbooru-tag-mcp is open source, so you can use and self-host it for free.

What are the best alternatives to danbooru-tag-mcp?

The closest alternatives to danbooru-tag-mcp are Best AI Image Prompt MCP, Stackmark, Meigen AI MCP, awesome-mcp-servers and MiniRAG.

Where this comes from

  • Description: written from the project's README and its MCP Registry entry.

How AIPoweredBot checks servers