Comfyui Agent Skill Mie
MieMieeeee/comfyui-agent-skill
Agent skill for running registered ComfyUI workflows through a stable CLI, and for importing a user's own ComfyUI workflow into their private registry after review.
Route ALL Anima image generation: validate Danbooru hard anchors, form visual brief, assemble English prompts and args, then load comfyui-manager for workflow execution.
$ npx skills add ShiroEirin/comfyui-good-anima --skill comfyui-animatool -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install ShiroEirin/comfyui-good-anima comfyui-animatool --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/ShiroEirin/comfyui-good-anima.git skills-src && mkdir -p .claude/skills && cp -r skills-src/comfyui-animatool .claude/skills/comfyui-animatool && rm -rf skills-srcUse ~/.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/
Install the "comfyui-animatool" agent skill from https://github.com/ShiroEirin/comfyui-good-anima/tree/main/comfyui-animatool into .claude/skills/comfyui-animatool/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "comfyui-animatool", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/ShiroEirin/comfyui-good-anima/tree/main/comfyui-animatoolType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add ShiroEirin/comfyui-good-anima --skill comfyui-animatool -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install ShiroEirin/comfyui-good-anima comfyui-animatool --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/ShiroEirin/comfyui-good-anima.git skills-src && mkdir -p .agents/skills && cp -r skills-src/comfyui-animatool .agents/skills/comfyui-animatool && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "comfyui-animatool" agent skill from https://github.com/ShiroEirin/comfyui-good-anima/tree/main/comfyui-animatool into .agents/skills/comfyui-animatool/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "comfyui-animatool", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add ShiroEirin/comfyui-good-anima --skill comfyui-animatool -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install ShiroEirin/comfyui-good-anima comfyui-animatool --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/ShiroEirin/comfyui-good-anima.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/comfyui-animatool .cursor/skills/comfyui-animatool && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "comfyui-animatool" agent skill from https://github.com/ShiroEirin/comfyui-good-anima/tree/main/comfyui-animatool into .cursor/skills/comfyui-animatool/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "comfyui-animatool", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/ShiroEirin/comfyui-good-anima.git --path comfyui-animatool--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add ShiroEirin/comfyui-good-anima --skill comfyui-animatool -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install ShiroEirin/comfyui-good-anima comfyui-animatool --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/ShiroEirin/comfyui-good-anima.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/comfyui-animatool .gemini/skills/comfyui-animatool && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "comfyui-animatool" agent skill from https://github.com/ShiroEirin/comfyui-good-anima/tree/main/comfyui-animatool into .gemini/skills/comfyui-animatool/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "comfyui-animatool", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install ShiroEirin/comfyui-good-anima comfyui-animatoolInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add ShiroEirin/comfyui-good-anima --skill comfyui-animatool -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/ShiroEirin/comfyui-good-anima.git skills-src && mkdir -p .github/skills && cp -r skills-src/comfyui-animatool .github/skills/comfyui-animatool && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "comfyui-animatool" agent skill from https://github.com/ShiroEirin/comfyui-good-anima/tree/main/comfyui-animatool into .github/skills/comfyui-animatool/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "comfyui-animatool", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add ShiroEirin/comfyui-good-anima --skill comfyui-animatool -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install ShiroEirin/comfyui-good-anima comfyui-animatool --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/ShiroEirin/comfyui-good-anima.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/comfyui-animatool .opencode/skills/comfyui-animatool && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "comfyui-animatool" agent skill from https://github.com/ShiroEirin/comfyui-good-anima/tree/main/comfyui-animatool into .opencode/skills/comfyui-animatool/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "comfyui-animatool", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
comfyui-animatoolRoute ALL Anima image generation: validate Danbooru hard anchors, form visual brief, assemble English prompts and args, then load comfyui-manager for workflow execution.
Comfyui Animatool is an agent skill from 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. Triggers: anima, comfyui, 生图, 画图, 出图, 生成, 来一张, roll, 抽卡, 画师融合.
Its SKILL.md is about 4.6k tokens, which your agent loads only when the skill is triggered. The skill folder holds 3 other files, including reference files (for example `references/artist-style-research.md` and `references/failure-patterns.md`).
It sits in AI & LLM Engineering, covering Diffusion and image models. It works with ComfyUI. The licence is GPL-3.0.
12 steps, taken from the step headings in SKILL.md.
Read from SKILL.md and the folder at commit 77fba84. It shows what the files ask for, not the result of running them.
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.
Shell commands in SKILL.md call:
nodeFrom the folder's file list and the shell code blocks in SKILL.md.
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Comfyui Animatool loads about 4.6k tokens when it runs, and up to ~5.6k if it reads all its reference files. Until then it costs about 63 tokens; SKILL.md has 1,352 words of instructions outside code blocks.
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.
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.
The full file from ShiroEirin/comfyui-good-anima at commit 77fba84, republished under its GPL-3.0 licence (© ShiroEirin). 1,352 words, ~4,585 tokens.
.claude/skills/comfyui-animatool/SKILL.md (or your agent's skills folder). This skill also uses 2 other files; get the full folder from GitHub.本 skill 是 Anima 生图的唯一入口。只做硬约束:路由、视觉简报、tag 校验、prompt 组装、冲突消解、args 输出。不替代模型构图常识;不执行 workflow。
danbooru-tags 校验后才回填。hard_tags、soft_phrases、nltags_block。danbooru-tags 只校验 hard_tags;不得校验或回填 soft_phrases。comfyui-manager。拿到用户需求后先判断分支:
| 用户意图 | 路线 |
|---|---|
| 普通 Anima 生图 | 视觉简报 → tag 校验 → 组装 prompt + args → 加载 comfyui-manager |
| 随机 / roll / 抽卡 | danbooru-tags --random → 筛选 → 视觉简报 → tag 校验 → prompt + args |
| 画师融合(明确说融合/混合/artist_chain) | artist_chain → 视觉简报 → prompt + args |
| 纯查 tag/画师/角色/作品 | 只用 danbooru-tags,不生图 |
| 只抽候选不生成 | danbooru-tags --random,不生图 |
"分别用 A/B 各出图"是多个普通 job,不是 Artist Mixer。 "允许多个画师"不是融合指令;每个非融合 job 仍选 1 个画师。
组装 prompt 前,先建立情境因果链,再拆解视觉简报字段。
发生了什么 → 角色的情感/欲望/冲突 → 具体反应(表情+肢体) → 环境如何参与 → 最抓人眼球的画面瞬间情境构建完成后,提取以下字段:
情境因果锁构建的 prompt 可能已经很有灵魂,但 tag 容易遗漏画面维度的某些方面。八维检查不是创作指导,而是查漏补缺:确认画面元素在八个维度上没有明显空缺。
| 维度 | 检查问题 | 缺失表现 | 补全方向 |
|---|---|---|---|
| 互动 | 元素之间有无行为联系? | 各自独立摆 pose,零交集 | 对视、触碰、动作呼应、人与环境互动、物体间物理关系 |
| 情感 | 表情+肢体传递了什么情绪? | generic smile/面无表情 | 微表情(surprised giggling, tender gaze)、身体语言(前倾/缩肩/攥拳) |
| 视线 | 目光或引导线指向哪里? | 所有人看镜头或闭眼 | 角色间对视、偷瞄、看向画外某物、视线跟随、构图引导线 |
| 联动 | 环境是否影响主体? | 环境是纯背景装饰 | 风雨 → 反应、光线 → 塑形、季节 → 细节、材质受环境影响(打湿/反光/积雪) |
| 动势 | 冻结画面暗示了运动吗? | 像摆拍立绘,重心正中 | 重心偏移、布料飞扬方向、头发飘动、飞溅/飘落轨迹、失衡感 |
| 空间 | 有前后层次和呼吸感吗? | 平铺直叙,贴脸输出 | 前景遮挡、景深虚化、正负空间、引导线、画面重心偏移 |
| 质感 | 材质有真实细节吗? | 塑料感/卡通化 | 湿润反光、粗糙纹理、丝滑垂坠、水珠凝结、皮肤毛孔 |
| 因果 | 观众能看出前因后果吗? | 不知道在发生什么 | 行为起因 → 当前姿态 → 暗示后续(即使是涩图:下流动作为因 → 表情/姿态为果) |
hard_tags 维持硬锚点干净。hard_tags、soft_phrases、nltags_block 是否语义分离。画布必须匹配构图。不默认竖图/方图/超宽。
| 比例 | 画布 | 用途 |
|---|---|---|
| 2:3 | 1024×1536 | 单人全身、立绘、手机壁纸 |
| 3:4 | 1152×1536 | 角色为主、轻环境 |
| 1:1 | 1024×1024 | 头像、半身、简单中心 |
| 1:1 | 1536×1536 | 复杂中心构图、方形海报 |
| 4:3 | 1536×1152 | 室内中景 |
| 3:2 | 1536×1024 | 多人互动、横向动作 |
| 16:9 | 1536×864 | 宽银幕、远景 |
| 9:16 | 864×1536 | 手机海报、竖向空间 |
| 2:1 | 1536×768 | 超宽场景 |
| 1:2 | 768×1536 | 超竖图 |
先选构图比例,再选分辨率;默认使用 1536 级画布。多人、复杂互动、大场景、强光影或脸部易糊时,可在同一比例上升到 2048 级;不得为了默认高清强行升分辨率。
先写检索计划(最多 4 个语义锚点),再调用 danbooru-tags 批量查询。典型生图最多 1 次批量查询 + 1 次补查。
检索顺序:
禁止把候选池作为默认检索入口。禁止为了普通描述批量扩展候选池。
confirmed_tags 只接受 exact_tag、exact_alias、artist prefix;candidate_tags 不直接回填。
必查:角色、作品/IP、最终选定画师。 可查:用户明确指定的服装/道具/姿势/概念锚点。 不查:构图、光影、氛围、情绪、连续动作 → 写 nltags。
confirmed_tags 按用户意图筛选后回填。candidate_tags 只作候选,不整批塞进 prompt。missing → 写 nltags,不伪造 Danbooru tag。@。danbooru-tags 校验。hard_tags(quality/meta/year/safety → count → character → series → artist → confirmed appearance/tags) → soft_phrases → nltags_blockhard_tags、soft_phrases、nltags_block,最后再拼成 prompt_11。prompt_11 = hard_tags + ", " + soft_phrases + ", " + nltags_block。hard_tags:经 danbooru-tags confirmed 或 Anima 固定控制词确认的离散 tag。soft_phrases:模型根据情境因果生成的短视觉短语,不走 danbooru-tags,不作为 Danbooru hard anchor。nltags_block:有语法结构的连续描述,负责空间、动作归属、接触、视线、遮挡、光源、景深、色彩和因果后果。hard_tags 必须位于最前,soft_phrases 居中,nltags_block 位于末尾。hard_tags。confirmed_tags。nltags_block。masterpiece, very aesthetic, best quality, score_9, score_8, highres, absurdres, newest, year 2025, nsfw裸模型/对比测试:
masterpiece, best quality, score_7, safe安全标签:safe / sensitive / nsfw / explicit。用户未指定时默认 nsfw。
负向按画面风险动态组装。不写 artist name。
核心:
worst quality, low quality, score_1, score_2, score_3, watermark, logo默认身体保护:
bad anatomy, bad hands, bad feet, extra fingers, missing fingers, distorted face, blurry按画面追加:
| 场景 | 追加 |
|---|---|
| 普通单人 | 默认身体保护 |
| 头像/半身/表情重点 | bad eyes, asymmetrical eyes, deformed face, blurry face |
| 全身/立绘 | extra limbs, missing limbs, disconnected limbs, bad feet |
| 动态动作/战斗/跳跃 | extra limbs, missing limbs, broken joints, disconnected limbs, bad hands, bad feet |
| 极端透视/低机位/俯视 | distorted face, bad perspective, broken joints, extra limbs |
| 手部特写/持物/道具互动 | bad hands, fused fingers, fused hands, extra fingers, missing fingers, malformed hands |
| 双人近距离互动 | merged bodies, extra arms, extra hands, cloned face |
| 多角色(3+) | duplicate, twins, merged bodies, fused limbs, extra limbs, cloned face, same outfit |
| 复杂服装/飘带/长发 | tangled hair, fused fabric, merged clothing, broken accessories |
| 背景虚化但主体清楚 | blurry face, blurry subject, out of focus face |
| 文字不是画面目标 | text |
规则:
blurry,改用 blurry face, blurry subject, out of focus face。text;保留 watermark, logo。@artist name(不加 @ 效果极弱)。artist_chain 不带 @,prompt_11 不重复画师名。例:wlop, (sakimichan:1.2), (krenz:0.7)。@artist。只在用户明确说"融合/混合/artist_chain/多画师合一"时启用。"分别用 A 和 B 各出图"是多个普通 job。
artist_chain 规则:
@,逗号或换行分隔。(name:1.2) 或 ::name::1.2。1.0,辅画师 0.2–0.4。默认 mixer 参数由 workflow 提供,不在 args 中传:
combine_mode=output_avg,fusion_mode=interpolate,artist_mixer_strength=1.0。artist_mixer_apply_to_uncond 默认 false,不要主动打开。newest, year 2025。newest,匹配 period tag。year 和 period 不走 danbooru-tags 校验。red hair,不写 red_hair。score_9 等质量 token 保留下划线。series 只写一次,不重复。character 和 series 都写。查不到独立 series tag 时只写 character。ye-pop、deviantart 等 dataset tag 只在用户明确要求非纯 anime/欧美插画风格时使用;默认 Anima 生图不加入。hard_tags。soft_phrases,不查 Danbooru。sitting, standing, wariza;复杂姿态因果用 nltags_block。nltags_block。nltags_block 不写离散 tag 列表;不写世界观解释、心理活动、纯文学比喻。hard_tags 不写完整英文句子。界限:确认标签 → hard_tags;视觉短语 → soft_phrases;连续空间-叙事描述 → nltags_block。
score_9、newest、year 2025 等horsing around, having fun, surprised giggling, grinning broadlystrong wind, cherry blossom blizzard, petals filling the airconfirmed_tags。Place X on the left...Use a close-up two-shot focusing on...as a sudden gust of wind sends petals into her face...Bright spring palette: vivid pink, sky blue pastel...约束:
hard_tags;不能确认但能增强画面的短语进 soft_phrases。nltags_block。nltags_block 不重复已在 hard_tags 中出现的外观/服装。nltags_block 按需要组织,不硬性限制句数,避免冗余。组装前必须消解以下冲突,逐项通过后才输出 args:
| 冲突对 | 规则 |
|---|---|
solo vs 多人 | 选一个,不共存 |
close-up vs full body | 选一个景别 |
from above vs from below | 选一个视角 |
from front vs from behind | 选一个朝向 |
closed eyes vs looking at viewer | 选一个视线 |
| 裸体 vs 服装 | 选一个着装状态 |
| 多角色属性归属 | 发色/服装必须绑定具体角色,不串 |
| 室内光源 vs 室外背景 | 光源和背景必须同空间 |
| 背光 | 必须补脸部补光或轮廓保护 |
| 宽景 vs 表情细节 | wide shot 不要求表情细节 |
单人正面默认保护脸部:保留 looking at viewer 或 facing viewer,nltags 补一句脸部清晰。
多人必须绑定:Place X on the front left, with [hair], [outfit], [action].
三层语义重复:同一概念只放一处。
(tag:2) 级别开始。Place / Use / Keep / Frame / Light / Blur。hard_tags 的角色、作品、画师、外观、服装硬锚点。单人示例:
Place the main subject slightly right of center.
Use soft window light from the left side.
Keep her face sharp and readable, with a softly blurred background.多人示例:
Place Reimu on the front left, with brown hair and a red shrine outfit.
Place Marisa on the front right, with blonde hair and a black witch dress.
Keep their hands separated and both faces readable.batch_size=1。batch_size=N。必须同时输出 workflow_id 和 args。
workflow_id 通过命令行传 run_workflow_args.js,不写进 args 文件。{"args":{...}} 嵌套。local/anima-txt2img-aesthetic-loralocal/anima-txt2img-aesthetic-lora-artist-mixerlocal/anima-txt2img-baseprompt_11:正向 promptprompt_12:负向 promptwidth、heightbatch_sizesteps:默认 30;高质量/复杂背景/多人/强光影用 40seed:可省略;省略时 comfyui-manager/workspace/run_workflow_args.js 会自动生成 1~4294967295 的随机整数并写回实际 args。用户指定则保留原值;重绘且未要求换 seed 时保留原 seed;用户要求固定复现时必须写入明确 seed。rtx_vsr_quality:默认 ULTRAfilename_prefix:anima/%year%-%month%-%day%/anima_base_v1_0-<artist|none>-<subject>采样器/调度器/CFG 由 workflow 提供,不在 args 中显式传。默认 workflow 使用 dpmpp_2m_sde_gpu + beta57 + CFG 4.5。
fls_*、teacache_*、anima_booster_*(除非用户明确要求调速/排障)rtx_vsr_scale逐项确认:
workflow_idargs(含 prompt_11、prompt_12、width、height、batch_size、steps、rtx_vsr_quality、filename_prefix)hard_tags、soft_phrases、nltags_block 已分离,nltags 位于 prompt_11 末尾全部通过后,执行:
node run_workflow_args.js submit <workflow_id> <args_json_file>submit 非阻塞;用户要求看结果时才查状态。
用户输入: "生成天使心跳的立华奏,三无感,教室窗边柔光"
分支: 普通生图。
视觉简报: Kanade Tachibana / solo / 教室窗边 / upper body eye-level / 1152×1536 / 左侧窗光柔光 / nltags: Place Kanade by the classroom window. Use soft window light from the left. Keep her face expressionless and readable with a softly blurred background.
Tag 校验: character=kanade tachibana + series=angel beats! 批量查 → confirmed。外观锚点(热门角色,模型已知):silver hair, yellow eyes, short hair, school uniform。
负向判断: 普通单人 → 追加全量:bad anatomy, bad hands, bad feet, extra fingers, missing fingers, distorted face, blurry。
冲突检查: solo ✓ / upper body 无冲突 ✓ / 窗光+教室同空间 ✓ / tag 与 nltags 不重复 ✓
prompt_11: masterpiece, very aesthetic, best quality, score_9, score_8, highres, absurdres, newest, year 2025, nsfw, 1girl, kanade tachibana, angel beats!, silver hair, yellow eyes, short hair, school uniform, classroom, window, looking at viewer, Place Kanade by the classroom window. Use soft window light from the left. Keep her face expressionless and readable with a softly blurred background.
prompt_12: worst quality, low quality, score_1, score_2, score_3, blurry, bad anatomy, bad hands, bad feet, extra fingers, missing fingers, distorted face, text, watermark, logo
Args: {"prompt_11":"...","prompt_12":"...","width":1152,"height":1536,"batch_size":1,"steps":30,"rtx_vsr_quality":"ULTRA","filename_prefix":"anima/%year%-%month%-%day%/anima_base_v1_0-none-kanade_tachibana"}(seed 省略时由 run_workflow_args.js 自动补随机整数)
workflow_id: local/anima-txt2img-aesthetic-lora → 执行 comfyui-manager submit。
用户输入: "用 @rella 画爱丽丝和魔理沙,春天幻想乡,强风樱花,CP 感"
hard_tags:masterpiece, very aesthetic, best quality, score_9, score_8, highres, absurdres, newest, year 2025, safe, 2girls, alice margatroid, kirisame marisa, touhou, @rella, blonde hair, blue eyes, green eyes, witch hat, shanghai doll, broomsoft_phrases:spring, cherry blossom blizzard, strong wind, having fun, surprised giggling, playful grin, pink petals filling the airnltags_block:Use a close-up two-shot focused on their faces and upper bodies. Place Alice on the left as petals hit her face, one hand raised to shield herself. Place Marisa on the right, holding her tilted witch hat while brushing a petal from Alice's cheek. Keep both faces sharp against a shallow blur of flying pink petals.用户输入: "融合 wlop 和 sakimichan,画原创魔法少女"
workflow_id:local/anima-txt2img-aesthetic-lora-artist-mixerartist_chain:wlop, (sakimichan:0.35)prompt_11 不写 @wlop 或 @sakimichan。hard_tags → soft_phrases → nltags_block 组装。用户输入: "随机一个画师来一张"
danbooru-tags --random 5 --for-prompt --json --compact。random_artists_for_prompt。只在遇到对应情况时读取,不自动加载:
| 场景 | 读取 |
|---|---|
| 用户要求画师背景、画风特色、构图倾向、"查底细" | references/artist-style-research.md |
| 生图结果出现人脸变形、切线粘连、比例失调、肢体归属混乱等 Anima 特有失败 | references/failure-patterns.md |
| 需要 tag 查询详细策略 | danbooru-tags/SKILL.md |
| 需要执行 workflow、排障、管理服务器 | comfyui-manager/SKILL.md |
© 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
SKILL.md and 2 other files (references) in comfyui-animatool of ShiroEirin/comfyui-good-anima.
Open the folder on GitHubat commit 77fba84
Comfyui Animatool 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.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Comfyui Animatool this skillShiroEirin/comfyui-good-anima | 481 | — | ~4.6k | Automated safety check: Pass | GPL-3.0 | |
| Comfyui Agent Skill MieMieMieeeee/comfyui-agent-skill | 116 | — | ~3.9k | Automated safety check: Pass | Apache-2.0 | |
| Headroommomori777/Artemis | 378 | — | ~562 | Automated safety check: Pass | MIT | |
| Comfyui Media Workflowscalesthio/generative-media-skills | 193 | — | ~8.5k | Automated safety check: Pass | MIT | |
| Comfyui Coreartokun/comfyui-mcp | 800 | — | ~3.3k | Automated safety check: Pass | MIT | |
| Comfyui Skill OpenclawHuangYuChuh/ComfyUI_Skills_OpenClaw | 413 | — | ~2.7k | Automated safety check: Pass | Apache-2.0 |
MieMieeeee/comfyui-agent-skill
Agent skill for running registered ComfyUI workflows through a stable CLI, and for importing a user's own ComfyUI workflow into their private registry after review.
momori777/Artemis
SmartCrusher + CCR context compression — crunch large JSON arrays, tool outputs, and search results to save tokens.
calesthio/generative-media-skills
Provider-independent production workflow for agents assembling, auditing, executing, and handing off ComfyUI node-graph workflows for image, video, upscale, inpaint, conditioning, and batch media…
artokun/comfyui-mcp
Core ComfyUI knowledge covering workflow format, node types, pipeline patterns, and MCP tool usage
HuangYuChuh/ComfyUI_Skills_OpenClaw
Run registered ComfyUI workflows through the fast comfyui-skill CLI, and use the official local Comfy MCP for live template, node, model, validation, and orchestration capabilities.
ConstantineB6/comfy-pilot
Helps an agent write ComfyUI custom nodes in Python, including wrapping an existing script, mapping data types and handling image batches.
ShiroEirin/comfyui-good-anima
A skill your agent uses for Anima / ComfyUI-AnimaTool image generation.
ShiroEirin/comfyui-good-anima
Search and validate Anima-compatible Danbooru tags, artists, characters, series, appearance, clothing, pose, scene, lighting, and prompt anchors.
ShiroEirin/comfyui-good-anima
Convert Anima image-generation intent, tags, references, or rough prompts into concrete composition decisions.
ShiroEirin/comfyui-good-anima
Generate randomized but Anima-compliant image parameters after comfyui-animatool confirms random-image generation intent.
ShiroEirin/comfyui-good-anima
Execute prepared ComfyUI workflows — submit/run/validate Anima args, check server/model/deps, manage queue.
ShiroEirin/comfyui-good-anima
Validate Danbooru hard anchors for Anima generation: artists, characters, series/IP, appearance, clothing, props, poses, expressions, scenes, and random candidates.
Works with
Categories
Route ALL Anima image generation: validate Danbooru hard anchors, form visual brief, assemble English prompts and args, then load comfyui-manager for workflow execution. Comfyui Animatool is an agent skill from 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.
Comfyui Animatool fits situations like: tasks that involve Diffusion and image models.
Run `npx skills add ShiroEirin/comfyui-good-anima --skill comfyui-animatool -a claude-code`. Or copy the skill folder (comfyui-animatool in ShiroEirin/comfyui-good-anima) into .claude/skills/comfyui-animatool in your project. Claude Code loads it when a task matches its description.
Run `npx skills add ShiroEirin/comfyui-good-anima --skill comfyui-animatool -a codex`. Or copy the skill folder (comfyui-animatool in ShiroEirin/comfyui-good-anima) into .agents/skills/comfyui-animatool in your project. Codex loads it when a task matches its description.
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 comfyui-animatool -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/comfyui-animatool, .gemini/skills/comfyui-animatool, .github/skills/comfyui-animatool and .opencode/skills/comfyui-animatool in your project.
Going by SKILL.md and its folder, Comfyui Animatool needs the command-line tools its instructions call (node).
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.
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.
Comfyui Animatool 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.
About 4.6k tokens (SKILL.md is roughly 18k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 1.1k tokens, read only when the agent opens those files.
Skills that share tags, products or a category with Comfyui Animatool: Comfyui Agent Skill Mie (MieMieeeee/comfyui-agent-skill, 116 stars), Headroom (momori777/Artemis, 378 stars), Comfyui Media Workflows (calesthio/generative-media-skills, 193 stars) and Comfyui Core (artokun/comfyui-mcp, 800 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
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.