nodes/Moeblack-ComfyUI-AnimaTool
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
33 Commits 3 Branches 0 Tags
dev
Commit Graph
13 Commits
Author SHA1 Message Date
Nakano Kenji d720b8f927 docs: 完善 QwenEdit 双引擎支持的指南与声明 2026-02-19 00:59:56 +08:00
Nakano Kenji e572748695 feat(mcp): 生图工具返回本地保存路径 
## 改动
- 在 \generate_anima_image\ 和 \generate_qwen_image\ 相关的响应中，附加图片在本地的绝对路径。

## 原因
方便大语言模型（如 QwenEdit）在多轮对话修图时，可以直接获取并引用上一轮生成的本地原图路径作为 \ase_image\。
 2026-02-18 23:41:01 +08:00
Nakano Kenji 6247d6b018 feat(qwen): 新增通义千问(Qwen)图像生成能力支持 
## 改动
- 新增 qwen_executor.py 与对应模板
- mcp_server 增加 generate_qwen_image 接口
- 更新依赖(uv.lock/pyproject.toml)
 2026-02-18 23:36:59 +08:00
Nakano Kenji 47987534e9 enhance: improve tool descriptions and history robustness 
- mcp_server: add explicit required/optional markers to all tool params, improve reroll schema to auto-convert required to optional-override, update default neg prompt

- history: fix next_id to scan global max across all records, add file fallback search when record not found in memory
 2026-02-07 06:26:00 +08:00
Nakano Kenji b71aa81721 refactor(schema): 精简 TOOL_SCHEMA description，详细规则移至 Skill/Wiki 
- artist/loras/models 描述大幅精简，只保留最小必要信息

- CURSOR_SKILL: 补充画师格式规则、LoRA 技术细节、reroll/批量/模型切换

- wiki/Prompt-Guide: 补充画师格式规则、LoRA 技术细节、reroll 与批量生成

- 新增 unet_name/clip_name/vae_name 可选参数暴露模型切换能力
 2026-02-06 18:33:35 +08:00
Nakano Kenji c743ef5981 docs: 更新文档覆盖 reroll/history/repeat/batch_size 新功能 
- README: Features、Parameters 表、HTTP 路由表、目录结构

- wiki/API-Reference: 新增 list_anima_history/reroll_anima_image/list_anima_models 文档

- wiki/Home: Features 列表补充

- wiki/FAQ: 更新批量生成答案，新增 reroll FAQ

- mcp_server: generate/reroll 响应追加历史 ID 提示

- anima_executor: generate 返回 history_id 字段
 2026-02-06 18:25:27 +08:00
Nakano Kenji a94afd0818 feat(reroll): 新增生成历史记录系统和 reroll/history MCP 工具 
- 新增 executor/history.py：HistoryManager（内存 deque + JSONL 持久化）

- generate 完成后自动记录历史（参数、seed、尺寸等）

- 新增 MCP 工具 list_anima_history 和 reroll_anima_image

- generate 和 reroll 均支持 repeat（多次 queue）和 batch_size

- HTTP server 新增 /history 和 /reroll 端点
 2026-02-06 17:57:31 +08:00
Nakano Kenji 6eb5fecd37 fix(tools): 修改了工具介绍 2026-02-06 12:01:34 +08:00
Nakano Kenji b240bc1318 fix(lora): clarify and normalize lora_name matching 2026-02-04 18:34:13 +08:00
Nakano Kenji c1c2e3dc0d Add parallel call support note to tool descriptions 2026-02-04 01:30:49 +08:00
Nakano Kenji d31d284ad5 fix: change round_to default from 8 to 16 for Cosmos spatial_patch_size compatibility 
Anima model (based on Cosmos architecture) requires latent space dimensions to be divisible by spatial_patch_size=2. Since VAE scales down by 8x, pixel dimensions must be divisible by 8*2=16.

Changes:

- Update round_to default from 8 to 16 in config.py and anima_executor.py

- Add align_dimension() function to align user-provided width/height

- Apply alignment to directly specified width/height (not just aspect_ratio)

- Update schema descriptions to document 16-multiple requirement
 2026-02-03 21:24:27 +08:00
Nakano Kenji b8682d9b5d Clarify multi-artist support: works but less stable, AI should prefer single artist 2026-02-03 17:12:39 +08:00
Nakano Kenji 7917b066f2 Initial release v1.0.0 
- MCP Server: Native image display in Cursor/Claude
- ComfyUI HTTP API: /anima/health, /anima/schema, /anima/knowledge, /anima/generate
- Standalone FastAPI server and CLI tool
- Structured prompt system with 14 aspect ratio presets
- Expert knowledge base for Anima prompting
 2026-02-03 16:18:57 +08:00