danbooru-tag-mcp
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@danbooru-tag-mcplook up the real danbooru tags for blue hair and maid dress before I write my prompt"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
danbooru-tag-mcp
A Danbooru tag lookup MCP (Model Context Protocol) server for local AI image generation: your AI client resolves real Danbooru tags — exact names, aliases, categories, post counts, co-occurrence statistics, wiki descriptions, and the tag lists of real posts — before composing prompts, which noticeably improves tag accuracy in generated images.
Quick start
Install per-user (Windows 10/11, no admin rights):
iwr -useb "https://raw.githubusercontent.com/BaixuanZhu/danbooru-tag-mcp/main/install.ps1" | iexPoint your MCP client at the command:
{
"mcpServers": {
"danbooru-tags": {
"command": "danbooru-tag-mcp",
"args": []
}
}
}That's it — the client picks up the seven tools below on its next session.
Related MCP server: Civitai MCP Server
Features
7 MCP tools: tag search, exact tag info, related tags, alias resolution, wiki lookup, post search, and tag distribution profiles
Tag distribution profiles:
get_tag_profilesamples a random page of a big tag's posts and answers what it actually renders as — co-occurring tags by in-sample frequency (corpus constants like1girl/solosplit intoubiquitous), plus characteristic combinations ranked by lift, so a 300k-post tag likearm_upcan be checked against what you actually wanted before it pollutes a promptTag lists back from post search:
search_postsreturns each post's full tag list split by category (copyright/artist/character/general/meta), so mining a few posts reveals which tags naturally go togetherTagger-era tag detection:
get_tag_infoclassifies every name against the pinned WD14 tagger vocabularies (wd14field:live/tagger_era— a tagger-era name Danbooru has since renamed, tryget_tag_alias/unknown), so LoRA card captions stop dead-ending at "tag not found";search_tagsaddswd14_hitsfallback hits when no result carries postsUncensored by default:
search_postsapplies no rating filter — results reflect a tag's whole population (arm_up is ~95% g/s/q); add arating:g/s/q/emetatag when you want only R-18 or only all-agesSelf-updating:
danbooru-tag-mcp upgradechecks GitHub Releases, verifies SHA256 and replaces the binary atomicallyZero-config install: per-user installer (setup wizard or one-line PowerShell) that registers the install directory into the user PATH automatically
Polite API usage: requests are throttled to one per 1.5 seconds; anonymous access works, credentials raise the limits (free/anonymous post searches are still capped at 2 content tags per query)
Install
Requires Windows 10/11 on x64 or ARM64.
Option 1: setup wizard (recommended)
Download danbooru-tag-mcp-windows-<arch>-setup.exe from the
latest release
and double-click it. This installs per-user to
%LOCALAPPDATA%\Programs\danbooru-tag-mcp (no admin rights needed) and
registers the user PATH.
Option 2: one-line PowerShell
iwr -useb "https://raw.githubusercontent.com/BaixuanZhu/danbooru-tag-mcp/main/install.ps1" | iexCustom install directory:
.\install.ps1 -InstallDir "D:\tools\danbooru-tag-mcp"Option 3: portable zip
Download danbooru-tag-mcp-windows-<arch>.zip from the
release page,
extract danbooru-tag-mcp.exe anywhere, and run it once — the first run
silently adds its own directory to the user PATH (idempotent).
Configure your MCP client
Point your MCP client (Claude Desktop, ZCode, etc.) at the command:
{
"mcpServers": {
"danbooru-tags": {
"command": "danbooru-tag-mcp",
"args": []
}
}
}The bare command name works because the install directory is on the user PATH; alternatively use the full exe path.
Tools
Tool | Description | Parameters |
| Search tags by keyword, ordered by post count; when no result has posts, adds |
|
| Exact info for one tag (category, post count), plus a |
|
| Tags that co-occur with the given tag (meta-category tags like |
|
| Resolve an alias or misspelling to its canonical tag ( |
|
| Wiki page of a tag: description, multilingual names, and the |
|
| Search posts by a tag combination, no rating filter by default (add a |
|
| Distribution profile of one tag from a random post sample (default 200 posts, whole population — no rating filter): |
|
CLI
danbooru-tag-mcp run the MCP server (what MCP clients launch)
danbooru-tag-mcp upgrade self-update via GitHub Release (SHA256 verified)
danbooru-tag-mcp version print the version
danbooru-tag-mcp help show helpEnvironment variables
DANBOORU_LOGIN— your Danbooru account name (the login you use on the site). Optional.DANBOORU_API_KEY— your Danbooru API key, generated from the API key section of your Danbooru profile. Optional. When bothDANBOORU_LOGINandDANBOORU_API_KEYare set, every request is sent authenticated with your account (higher rate limits and content access); otherwise the client is anonymous and Danbooru applies lower limits.DANBOORU_MCP_NO_BOOTSTRAP— skip the startup PATH registration when non-empty.
Building from source
Requires Go 1.26+ and GNU Make (bundled with Git Bash):
make test # unit tests
make integration # online smoke test against the live Danbooru API (~1 min)
make build # dev build -> dist/<arch>/ (bootstrap disabled)
make release # portable zip + checksums.txt
make installer # NSIS setup exe (requires makensis: scoop install nsis)Releases are automated: pushing a v* tag makes GitHub Actions run the test
suite and publish the setup exes, portable zips, and checksums.txt for both
architectures.
This server cannot be deployed
Maintenance
Related MCP Connectors
Free public web freshness and response-metadata checks for AI agents.
Free verified network knowledge for AI agents: deterministic answers, honest unknowns.
Access and interact with anime and manga data seamlessly. Retrieve detailed information about your…
Serves your design system and coding standards to coding agents, so they stop guessing.
Related MCP Servers
- FlicenseNot gradedqualityFmaintenanceEnables searching and analyzing Danbooru tags, character statistics, and wiki information. Extracts tags from posts, analyzes frequently used tags for characters with customizable filters, and retrieves detailed wiki entries.1-
- AlicenseNot gradedqualityDmaintenanceEnables interaction with the Civitai API to search for AI models, browse images, and access creator information. It provides tools for filtering models by popularity, rating, or type and retrieving detailed version metadata.MIT
- AlicenseAqualityDmaintenanceA Python MCP server that wraps the Gelbooru API. Connect it to any MCP-compatible client (Claude Desktop, Cursor, etc.) to search posts, look up tags, and generate Stable Diffusion prompts from real character appearance data — all directly from your AI assistant.871MIT
- AlicenseAqualityCmaintenanceProvides AI agents with direct access to local Yomitan dictionary databases for offline Japanese vocabulary lookups, kanji searches, and sentence tokenization. It leverages the Yomitan browser extension's API to enable rich dictionary interactions and Anki flashcard field generation.59 npm6Mozilla Public 2.0