Agent skill

Danbooru Tags

by ShiroEirin in ShiroEirin/comfyui-good-anima

Validate Danbooru hard anchors for Anima generation: artists, characters, series/IP, appearance, clothing, props, poses, expressions, scenes, and random candidates.

GPL-3.0Auto-check passed

Install Danbooru Tags

skills CLI
$ npx skills add ShiroEirin/comfyui-good-anima --skill danbooru-tags -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install ShiroEirin/comfyui-good-anima danbooru-tags --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/ShiroEirin/comfyui-good-anima.git skills-src && mkdir -p .claude/skills && cp -r skills-src/danbooru-tags .claude/skills/danbooru-tags && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
danbooru-tags
GitHub stars
481
Token cost
~1.5k tokens
SKILL.md length
314 words
Files
24
Skills in repo
8
Repo updated
First seen
Licence
GPL-3.0

At a glance

Validate Danbooru hard anchors for Anima generation: artists, characters, series/IP, appearance, clothing, props, poses, expressions, scenes, and random candidates.

  • Works in 5 steps: 精确验证:先用模型已知、用户给定或已解析的 canonical tag 查询。 → 小范围补查:精确验证无命中时,只补同一语义锚点的别名、英文名、拆分词。 → 候选池:补查仍无命中,且该锚点必须落成 tag 时,才取同 group 候选。 → …
  • SKILL.md covers 硬约束, CLI 位置, 检索优先级 and 批量查询(生图前默认), plus 8 more sections
  • Runs Rust and Python scripts from its folder

What it does

Danbooru Tags is an agent skill from ShiroEirin/comfyui-good-anima. Validate Danbooru hard anchors for Anima generation: artists, characters, series/IP, appearance, clothing, props, poses, expressions, scenes, and random candidates. Not for composition, prompt writing, or workflow execution.

Its SKILL.md is about 1.5k tokens, which your agent loads only when the skill is triggered. The skill folder holds 26 other files (for example `build_index.py`).

The licence is GPL-3.0.

Example prompts

  • “/danbooru-tags”

Requirements

  • Python 3

Workflow steps

5 steps, taken from the first numbered list in SKILL.md.

  1. 精确验证:先用模型已知、用户给定或已解析的 canonical tag 查询。
  2. 小范围补查:精确验证无命中时,只补同一语义锚点的别名、英文名、拆分词。
  3. 候选池:补查仍无命中,且该锚点必须落成 tag 时,才取同 group 候选。
  4. nltags:复合概念、关系、构图、光影、查不到的自然语言描述写入 nltags。
  5. 随机池:仅用户明确要求随机/roll/抽卡时调用。

What it can do on your machine

Read from SKILL.md and the folder at commit 77fba84. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships script files (Rust and Python, from the files we listed), which the agent can run.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Danbooru Tags loads about 1.5k tokens when it runs. Until then it costs about 60 tokens; SKILL.md has 314 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~60
When it runs · the whole SKILL.md, loaded when a task matches
~1.5k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from ShiroEirin/comfyui-good-anima at commit 77fba84, republished under its GPL-3.0 licence (© ShiroEirin). 314 words, ~1,494 tokens.

Download SKILL.mdSave it as .claude/skills/danbooru-tags/SKILL.md (or your agent's skills folder). This skill also uses 23 other files; get the full folder from GitHub.
name
danbooru-tags
description
Validate Danbooru hard anchors for Anima generation: artists, characters, series/IP, appearance, clothing, props, poses, expressions, scenes, and random candidates. Not for composition, prompt writing, or workflow execution.

danbooru-tags

纯标签检索工具。不决定构图、不写 prompt、不执行 workflow。

硬约束

  • 始终使用本地 bin/danbooru-tags.exe,不搜索、不递归。
  • 默认先用已知/推断出的 Danbooru canonical tag 精确查询。
  • 精确查询查不到时,才进入同组别名、拆分词、候选词补查。
  • 候选池和随机池不是默认入口;仅在用户要求随机/抽卡,或精确补查仍缺失时使用。
  • confirmed_tags 可用于回填(仍需意图筛选);candidate_tags 仅作候选,必须筛选。
  • missing → 写入 nltags。
  • 不决定构图、镜头、光影、workflow 执行。
  • 不写完整 prompt。
  • 非 artist category 的 tag 不加 @,不作为画师。
  • 随机角色/服装/姿势/场景候选不使用 --for-prompt。

CLI 位置

powershell
$SKILLS_ROOT = $env:COMFYUI_GOOD_ANIMA_SKILLS_DIR
if (-not $SKILLS_ROOT) { throw "COMFYUI_GOOD_ANIMA_SKILLS_DIR is required" }
$TAG_ROOT = Join-Path $SKILLS_ROOT "danbooru-tags"
$cli = Join-Path $TAG_ROOT "bin/danbooru-tags.exe"

不搜索递归。不使用旧路径。

检索优先级

按顺序执行:

  1. 精确验证:先用模型已知、用户给定或已解析的 canonical tag 查询。
    • 默认 --match-mode auto:先 exact,exact 无命中才 fuzzy。
    • 只允许直接命中时使用 --match-mode exact。
    • artist:--group artist --prefix "@artist name"
    • character:--group character --keyword "character name"
    • series/IP:--group series --keyword "series name"
    • general hard anchor:--group <group> --keyword "tag phrase"
  2. 小范围补查:精确验证无命中时,只补同一语义锚点的别名、英文名、拆分词。
  3. 候选池:补查仍无命中,且该锚点必须落成 tag 时,才取同 group 候选。
  4. nltags:复合概念、关系、构图、光影、查不到的自然语言描述写入 nltags。
  5. 随机池:仅用户明确要求随机/roll/抽卡时调用。

禁止:

  • 不先精确验证就直接查候选池。
  • 不把 group 候选整批塞进 prompt。
  • 不为普通描述扩展大候选池。
  • 不为同一锚点连续循环补查。

批量查询(生图前默认)

默认批量只放精确验证项和必要的小范围补查项。JSON 必须 UTF-8 without BOM:

powershell
$json = @{
  queries = @(
    @{ id = "character"; group = "character"; keyword = "kanade tachibana"; limit = 5 },
    @{ id = "series"; group = "series"; keyword = "angel beats"; limit = 5 },
    @{ id = "artist"; group = "artist"; prefix = "@mignon"; limit = 5 }
  )
} | ConvertTo-Json -Depth 20
$file = Join-Path $env:TEMP "danbooru_batch.json"
[System.IO.File]::WriteAllText($file, $json, [System.Text.UTF8Encoding]::new($false))
& $cli --batch-file $file --batch-workers 8 --for-prompt --json --compact

单查

powershell
# 画师
& $cli --group artist --prefix "@mignon" --limit 5 --for-prompt --json --compact

# 角色
& $cli --group character --keyword "hakurei reimu" --limit 5 --for-prompt --json --compact

# 只做直接命中验证,不返回模糊候选
& $cli --group character --keyword "hakurei reimu" --match-mode exact --limit 5 --for-prompt --json --compact

随机

意图命令输出字段数量
抽 N 候选画师挑 1--random N --jsonrandom_artistsN 条
角色/服装/姿势/场景候选--random N --group <group> --jsonrandom_tagsN 条
随机 1 画师直接生图回填--random 5 --for-prompt --json --compactrandom_artists_for_prompt1 条

规则:

  • 候选池用 --random N --json,生图回填用 --random N --for-prompt(只返回 1 条),不混用。
  • --for-prompt 不允许与 --group 一起用。
  • N 使用 10–50,上限 300。只调用一次,不循环。
  • 不向用户复述完整候选 JSON。

批量规则

  • 同一锚点优先放 canonical 主词;别名、拆分词只在主词不稳或无命中风险高时加入。
  • 精确 artist/character/series 查询 limit=5。
  • 服装/动作/场景/俗称 limit=10–20。
  • 普通生图 ≤4 语义锚点;稳定 canonical 每锚点 1 个 query,不稳定锚点最多 2–3 变体,总 query 4–12。
  • 最多 1 次批量 + 1 次补查。
  • group 漏匹配时同批加 category=general 变体。
  • general 命中仅作 candidate_tags。

查 vs 不查

查: 用户指定角色/作品/画师、关键服装/道具/姿势、随机候选、已解析 canonical 候选需确认。

不查(写 nltags): 普通自然语言描述、光源方向、前景/中景/背景、脸部清晰/故事关系/氛围。

输出读取

  • confirmed_tags — 可回填,仍需意图筛选。
  • confirmed_tags 只来自 exact_tag、exact_alias、artist prefix。
  • candidate_tags — 模糊补查或回退候选,必须筛选,不直接回填。
  • missing — 写 nltags。
  • 不向用户复述完整搜索过程。

JSON 输出 schema

单查 --for-prompt --json --compact
json
{
  "found": true,
  "confirmed_tags": {
    "characters": [
      {
        "tag": "hakurei_reimu",
        "prompt_tag": "hakurei reimu",
        "category": "characters",
        "source_category": "characters",
        "count": 65105,
        "match_score": 1000,
        "match_layer": "exact_tag"
      }
    ]
  },
  "candidate_tags": {}
}
批量 --batch-file ... --for-prompt --json --compact
json
{
  "found": true,
  "results": {
    "character": {
      "found": true,
      "confirmed_tags": {},
      "candidate_tags": {}
    }
  },
  "missing": [],
  "usage": {
    "confirmed_tags": "...",
    "candidate_tags": "...",
    "nltags_hint": "...",
    "empty_result": "..."
  }
}

字段规则:

  • found=false → 不回填 tag。
  • confirmed_tags.<category>[] → 可作为 hard tag 候选,仍按用户意图筛选。
  • candidate_tags.<category>[] → 只用于人工/模型筛选,不直接写入 hard_tags。
  • missing[] → 对应 query 没有可确认 tag;停止补查或改写为 nltags。
  • match_layer=exact_tag → tag 直接命中。
  • match_layer=exact_alias → alias 精确命中。
  • match_layer=prefix → artist prefix 命中。
  • match_layer=fuzzy / group_general_fallback → 候选,不是 confirmed。

硬性规则

  • Artist tag 必须来自 artist category,保留 @。
  • Character/series/general tag 不带 @,不作为画师。
  • confirmed_tags 不证明默认服装/发色稳定。
  • 不伪造 Danbooru tag。
  • CLI 不可用/非 0/非 JSON/字段缺失 → 停止,不切换旧实现。
  • newest/recent/mid/early/old 和 year XXXX 是 Anima 年代控制词,不需 tag 命中证明。

分组速查

group内容
artist画师,保留 @
character角色
series/ip/copyright作品
appearance/body发色、瞳色、发型、体型
expression表情
pose/action姿势、动作
clothing/outfit服装
accessory/prop配饰、道具
scene/background场景、天气
lighting光影 tag
metahighres 等

© ShiroEirin, GPL-3.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 23 other files in danbooru-tags of ShiroEirin/comfyui-good-anima.

  • SKILL.md
  • Anima-preview-alternate.csv
  • Anima-preview.csv
  • anima-1.0.csv
  • banned_tags.csv
  • bin/danbooru-tags.exe
  • build_index.py
  • rust-cli/Cargo.lock
  • rust-cli/Cargo.toml
  • rust-cli/src/cli.rs
  • rust-cli/src/db.rs
  • rust-cli/src/format.rs
  • rust-cli/src/groups.rs
  • rust-cli/src/lib.rs
  • rust-cli/src/main.rs
  • rust-cli/src/prompt.rs
  • rust-cli/src/result_item.rs
  • rust-cli/src/search.rs
  • … and 6 more

Open the folder on GitHubat commit 77fba84

Compare with similar skills

Danbooru Tags next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Danbooru Tags compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Danbooru Tags this skillShiroEirin/comfyui-good-anima481—~1.5kAutomated safety check: PassGPL-3.0
Seedance Character Consistency GuideEmily2040/seedance-2.07.6k1 repos~1.2kAutomated safety check: PassMIT
Anchor Textthedaviddias/Front-End-Checklist74k—~426Automated safety check: PassMIT
Character Riggingcalesthio/OpenMontage66k—~460Automated safety check: PassMIT
Review Tagshashicorp/terraform-provider-aws11k—~462Automated safety check: PassMPL-2.0
Hard Callalirezarezvani/claude-skills28k—~1.8kAutomated safety check: PassMIT

Similar skills

  • Keeps a recurring AI-video character's identity, wardrobe and expressions stable across shots with a stable tag, an identity anchor and one explicit action per character.

    7.6k GitHub starsUsed in 1 repo~1.2k tokens
    Media & CreativeAuto-check passed
  • Anchor Text

    thedaviddias/Front-End-Checklist

    A skill your agent uses when auditing metadata, crawlability, structured data, or indexability related to Use descriptive anchor text.

    74k GitHub stars~426 tokensUpdated 4 days ago
    Marketing & SEOAuto-check passed
  • Character Rigging

    calesthio/OpenMontage

    Build data-driven 2D character rigs for local animation: parts, pivots, layers, constraints, views, and reusable rig packages.

    66k GitHub stars~460 tokensUpdated 7 days ago
    Media & CreativeAuto-check passed
  • Review Tags

    hashicorp/terraform-provider-aws

    Official

    Review Terraform AWS Provider tagging: the tags/tagsall schema attributes, the @Tags(identifierAttribute=...) annotation, and wiring input.Tags = getTagsIn(ctx) after flex.Expand.

    11k GitHub stars~462 tokensUpdated today
    DevOps & CloudAuto-check passed
  • Hard Call

    alirezarezvani/claude-skills

    /em:hard-call — Framework for decisions with no good options.

    28k GitHub stars~1.8k tokensUpdated 1 mo ago
    Auto-check passed
  • Anchor Browser Automation

    ComposioHQ/awesome-claude-skills

    Automate Anchor Browser tasks via Rube MCP (Composio). An agent skill from ComposioHQ/awesome-claude-skills.

    77k GitHub starsUsed in 3 repos~757 tokens
    Productivity & AutomationAuto-check passed

More from ShiroEirin/comfyui-good-anima

All 8 skills in this repo
  • Comfyui Animatool

    ShiroEirin/comfyui-good-anima

    Route ALL Anima image generation: validate Danbooru hard anchors, form visual brief, assemble English prompts and args, then load comfyui-manager for workflow execution.

    481 GitHub stars~4.6k tokensUpdated 3 mo ago
    Auto-check passed
  • Comfyui Animatool

    ShiroEirin/comfyui-good-anima

    A skill your agent uses for Anima / ComfyUI-AnimaTool image generation.

    481 GitHub stars~3.4k tokensUpdated 3 mo ago
    Auto-check passed
  • Danbooru Tags

    ShiroEirin/comfyui-good-anima

    Search and validate Anima-compatible Danbooru tags, artists, characters, series, appearance, clothing, pose, scene, lighting, and prompt anchors.

    481 GitHub stars~2.4k tokensUpdated 3 mo ago
    Auto-check passed
  • Anima Composition Director

    ShiroEirin/comfyui-good-anima

    Convert Anima image-generation intent, tags, references, or rough prompts into concrete composition decisions.

    481 GitHub stars~2.9k tokensUpdated 3 mo ago
    Auto-check passed
  • Anima Random Gen

    ShiroEirin/comfyui-good-anima

    Generate randomized but Anima-compliant image parameters after comfyui-animatool confirms random-image generation intent.

    481 GitHub stars~1.2k tokensUpdated 3 mo ago
    Auto-check passed
  • Comfyui Manager

    ShiroEirin/comfyui-good-anima

    Execute prepared ComfyUI workflows — submit/run/validate Anima args, check server/model/deps, manage queue.

    481 GitHub stars~1.4k tokensUpdated 3 mo ago
    Auto-check passed

Questions about Danbooru Tags

What does Danbooru Tags do?

Validate Danbooru hard anchors for Anima generation: artists, characters, series/IP, appearance, clothing, props, poses, expressions, scenes, and random candidates. Danbooru Tags is an agent skill from ShiroEirin/comfyui-good-anima. Validate Danbooru hard anchors for Anima generation: artists, characters, series/IP, appearance, clothing, props, poses, expressions, scenes, and random candidates.

How do I install Danbooru Tags in Claude Code?

Run `npx skills add ShiroEirin/comfyui-good-anima --skill danbooru-tags -a claude-code`. Or copy the skill folder (danbooru-tags in ShiroEirin/comfyui-good-anima) into .claude/skills/danbooru-tags in your project. Claude Code loads it when a task matches its description.

How do I install Danbooru Tags in Codex?

Run `npx skills add ShiroEirin/comfyui-good-anima --skill danbooru-tags -a codex`. Or copy the skill folder (danbooru-tags in ShiroEirin/comfyui-good-anima) into .agents/skills/danbooru-tags in your project. Codex loads it when a task matches its description.

Can I use Danbooru Tags in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add ShiroEirin/comfyui-good-anima --skill danbooru-tags -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/danbooru-tags, .gemini/skills/danbooru-tags, .github/skills/danbooru-tags and .opencode/skills/danbooru-tags in your project.

What does Danbooru Tags need to run?

Going by SKILL.md and its folder, Danbooru Tags needs Rust and Python for the scripts in its folder. Our summary lists: Python 3.

Does Danbooru Tags access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Danbooru Tags safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Danbooru Tags use?

Danbooru Tags is published under the GPL-3.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Danbooru Tags use?

About 1.5k tokens (SKILL.md is roughly 6k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Danbooru Tags?

Skills that share tags, products or a category with Danbooru Tags: Seedance Character Consistency Guide (Emily2040/seedance-2.0, 7.6k stars), Anchor Text (thedaviddias/Front-End-Checklist, 74k stars), Character Rigging (calesthio/OpenMontage, 66k stars) and Review Tags (hashicorp/terraform-provider-aws, 11k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Danbooru Tags?

ShiroEirin (a GitHub user) maintains it in ShiroEirin/comfyui-good-anima, which has 481 GitHub stars. The repository holds 8 skills in this directory. The repository was last updated on June 15, 2026.

Source: ShiroEirin/comfyui-good-anima on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.