{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_cpqm69pgj4wp","handle":"tf-tools-mcp","url":"https://wellknown.network/agents/tf-tools-mcp","links":{"self":"https://wellknown.network/agents/tf-tools-mcp/record.json","html":"https://wellknown.network/agents/tf-tools-mcp","markdown":"https://wellknown.network/agents/tf-tools-mcp/record.md","api":"https://wellknown.network/api/v1/agents/tf-tools-mcp","status":"https://wellknown.network/api/v1/agents/tf-tools-mcp/status","claim":"https://wellknown.network/agents/tf-tools-mcp/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/tf-tools-mcp/claim.json","badge":"https://wellknown.network/agents/tf-tools-mcp/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/tf-tools-mcp/history","tools":"https://wellknown.network/api/v1/agents/tf-tools-mcp/tools"},"ard":{"identifier":"urn:air::server:tf-tools-mcp","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"tf-tools-mcp","summary":"Image generation and understanding tools using OpenRouter and Zhipu AI","description":"# TF Tools MCP\n\n基于 OpenRouter 和智谱 AI 的 MCP (Model Context Protocol) 图像生成与理解工具集。\n\n## 功能特性\n\n- **文本生成图像 (Text-to-Image)**: 使用多种模型通过 OpenRouter Images API 生成图像\n  - 支持模型：Gemini 3.1 Flash Image、Gemini 3 Pro Image、GPT Image 2/Mini、Seedream 4.5、FLUX.2 Max/Pro\n  - 可选的 AI 提示词增强功能（使用 GLM-4.6V）\n  - 多种质量级别和布局选项\n\n- **图像生成图像 (Image-to-Image)**: 使用文本引导转换参考图像\n  - 支持本地文件和 URL\n  - 针对不同转换模式的智能提示词增强\n\n- **图像理解 (Image Understanding)**: 使用 GLM-5V/4.6V 分析图像，标准 OCR 自动路由至 GLM-OCR\n  - 单图和批量图像分析\n  - 多种分析模板：描述、OCR、物体检测、图表分析\n  - 详细的 token 使用统计\n\n- **文件预览 (File Preview)**: 智能文件预览和摘要生成\n  - 支持 PDF、Word、Excel、PPT、图像等多种格式\n  - 自动提取元数据\n  - AI 驱动的内容摘要\n\n## 安装\n\n### 使用 uv（推荐）\n\n```bash\n# 克隆仓库\ngit clone https://github.com/A2C-SMCP/tf_tools_mcp.git\ncd tf_tools_mcp\n\n# 安装依赖\nuv sync\n\n# 开发模式安装\nuv pip install -e .\n```\n\n### 从 PyPI 安装\n\n```bash\npip install tf-tools-mcp\n```\n\n## 配置\n\n### 1. 环境变量配置\n\n创建 `.env` 文件并添加 API 密钥：\n\n```bash\n# 必需：图像生成\nOPENROUTER_API_KEY=你的_openrouter_api_key\n\n# 必需：图像理解和提示词增强\nZHIPUAI_API_KEY=你的_智谱ai_api_key\n\n# 可选设置\nIMAGE_STORAGE_DIR=./generated_images\nENABLE_PROMPT_ENHANCEMENT=true\n```\n\n### 2. 获取 API 密钥\n\n- **OpenRouter API Key**: 访问 [https://openrouter.ai/](https://openrouter.ai/) 注册，在 [https://openrouter.ai/keys](https://openrouter.ai/keys) 获取密钥\n- **智谱 AI API Key**: 访问 [https://open.bigmodel.cn/](https://open.bigmodel.cn/) 注册，在 [https://open.bigmodel.cn/usercenter/apikeys](https://open.bigmodel.cn/usercenter/apikeys) 获取密钥\n\n## MCP 客户端配置\n\n### Claude Desktop\n\n**macOS**: `~/Library/Application Support/Claude/claude_desktop_config.json`\n**Windows**: `%APPDATA%\\Claude\\claude_desktop_config.json`\n\n#### 方法 1：使用 uvx（推荐）\n\n无需手动安装，自动从 PyPI 运行：\n\n```json\n{\n  \"mcpServers\": {\n    \"tf-tools-mcp\": {\n      \"command\": \"uvx\",\n      \"args\": [\"tf-tools-mcp\"],\n      \"env\": {\n        \"OPENROUTER_API_KEY\": \"你的_api_key\",\n        \"ZHIPUAI_API_KEY\": \"你的_api_key\"\n      }\n    }\n  }\n}\n```\n\n#### 方法 2：使用已安装的包\n\n如果已通过 `pip install tf-tools-mcp` 安装：\n\n```json\n{\n  \"mcpServers\": {\n    \"tf-tools-mcp\": {\n      \"command\": \"tf-tools-mcp\",\n  …","publisher":{"name":"huruize","url":null},"homepage":"https://github.com/A2C-SMCP/tf_tools_mcp","repository":"https://github.com/A2C-SMCP/tf_tools_mcp","version":"0.1.7","license":"MIT","protocols":["mcp"],"tags":["mcp","image-generation","image-understanding","openrouter","zhipuai","glm"],"pricing":null,"endpoints":[{"url":"pypi:tf-tools-mcp","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","license":"pypi","repoUrl":"pypi","summary":"pypi","version":"pypi","description":"pypi","homepageUrl":"pypi","publisherName":"pypi"}},"derived":{"capabilities":[{"slug":"media.image-generation","name":"Image Generation","confidence":1,"provenance":"declared"},{"slug":"ai.model-access","name":"Model Access","confidence":1,"provenance":"declared"},{"slug":"media.image-understanding","name":"Image Understanding","confidence":1,"provenance":"declared"},{"slug":"dev.version-control","name":"Version Control","confidence":0.745,"provenance":"derived"}],"categories":["ai","dev","media"],"language":"en"},"observed":{"status":"unknown","statusReason":"Distributed as a package to run locally; no network endpoint to check.","lastOkAt":null,"lastProbedAt":null,"statusComputedAt":null,"reliability30d":null,"latestObservations":[],"tools":null,"package":{"name":"tf-tools-mcp","registry":"pypi","observedAt":"2026-10-22T22:24:24.114Z","publishedAt":"2026-07-31T02:49:13.103462Z","latestVersion":"0.1.7"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"tf-tools-mcp","url":"https://pypi.org/project/tf-tools-mcp/","firstSeenAt":"2026-09-10T14:22:12.659Z","fetchedAt":"2026-10-22T22:22:19.466Z","normalizedAt":"2026-10-22T22:22:19.466Z"}]},"firstSeenAt":"2026-09-10T14:22:12.659Z","updatedAt":"2026-10-22T22:24:24.114Z"}