character-recognize-mcp
Identifies characters and works in anime illustrations and looks up original Pixiv images and artists.
Click on "Install 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., "@character-recognize-mcpIdentify this anime character and find the Pixiv source: /home/user/image.png"
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.
Akashic 插画识别插件
在 Akashic 中识别动漫插画里的角色和作品,并反查 Pixiv 原图、画师或动画截图出处。
使用前准备
必须准备:
一个支持图片输入、兼容 OpenAI Chat Completions 接口的视觉模型 API Key。
默认使用阿里云百炼 DashScope 的
qwen-vl-max。
可选准备:
SauceNAO API Key。没有 Key 也能匿名反查,但额度更低。
Trace.moe 不需要配置 Key。
Related MCP server: ZeroChan MCP
安装
插件安装必须由 Akashic 当前正在运行的会话发起。请在对话中要求 Akashic 安装本插件;Akashic 会在当前 active turn 中执行:
.venv/bin/python main.py plugin-install \
--source https://github.com/lfegg/character-recognize-mcp.git \
--marketplace github安装完成后,终端会显示插件数据目录,通常类似:
<Akashic 工作目录>/plugin-data/character-recognize-github/安装命令完成后,安装操作所在的 Akashic turn 必须启动 attached programmatic child 验证候选版本。候选只向验证 child 暴露无网络、无凭据的 verify_installation 探针;探针成功后正常结束 child 和父 turn,Core 才会自动切换版本,下一 turn 生效。不要在 Akashic 会话之外直接运行安装命令,也不需要重启 Akashic。
配置
在插件数据目录中创建 config.local.toml:
request_timeout_seconds = 30
rate_limit_enabled = true
[vision]
base_url = "https://dashscope.aliyuncs.com/compatible-mode/v1"
api_key = "<你的 DASHSCOPE_API_KEY>"
model = "qwen-vl-max"
[saucenao]
api_key = "<你的 SAUCENAO_API_KEY>"vision.api_key 必须填写。没有 SauceNAO Key 时,删除 [saucenao] 两行即可使用匿名模式。
限制私密配置文件的访问权限:
chmod 600 /path/to/plugin-data/character-recognize-github/config.local.toml不要把 API Key 写入仓库、命令行参数或聊天消息。
配置完成后检查插件:
.venv/bin/python main.py plugin-doctor character-recognize@githubplugin-doctor 应返回 ok。如果报告 MCP runtime 缺失、source symlink 越界或候选无法启动,这是 Akashic Core 的安装/runtime 问题,不应通过删除插件安全检查或改写本插件启动命令绕过。
怎么用
先确保图片位于运行 Akashic 的电脑上。支持 JPEG、PNG、WebP 和 GIF,单张图片不能超过 20 MiB。
在 Akashic 对话中直接提供图片的绝对路径,例如:
识别这张插画里的角色和出处:/Users/me/Pictures/example.png也可以只查询其中一项:
这张图里是谁:/Users/me/Pictures/example.png帮我找这张图的 Pixiv 原图和画师:/Users/me/Pictures/example.png识别结果会区分“已命中”和“疑似候选”。置信度低于 0.7 的结果不会被当作确认答案;没有可靠来源时,插件不会编造 Pixiv 链接。
使用其他视觉模型
插件也支持其他 OpenAI compatible 图片模型。以 OpenCode Go 的 mimo-v2.5 为例,将 [vision] 替换为服务商提供的配置:
[vision]
base_url = "<服务商提供的 compatible API 地址>"
api_key = "<你的 API Key>"
model = "mimo-v2.5"接口地址必须使用 HTTPS,并且模型必须支持图片输入。
开发者需要查看 MCP 工具、返回结构、环境变量和测试方式时,请阅读 开发文档。
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
No tool schema history has been recorded yet.
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Connectors
Paid anime figure identification, market, rarity, authenticity and buying intelligence for AI agents
71Image risk scoring, EXIF, reverse-image backlinks, and image content detection via PicDefense.io.
Analyze images from multiple angles to extract detailed insights or quick summaries. Describe visu…
Generate, edit, and explore AI images. Flux, Imagen, LoRA identity swap, upscale, and more.
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-
- AlicenseAqualityDmaintenanceMCP server for the Zerochan anime image board API. Browse the global feed, search by tags, and retrieve full entry metadata — directly from any MCP-compatible AI assistant.3MIT
- AlicenseAqualityCmaintenanceSearches Danbooru for anime/game characters and their associated tags, providing tools to retrieve character profiles, co-occurring tags, wiki descriptions, and tag implications.512MIT
- AlicenseAqualityCmaintenanceAnalyzes images, audio, and video using Google's Gemini AI, with support for downloading videos from URLs such as Bilibili and YouTube.3MIT
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/lfegg/character-recognize-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server