{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_69t9gqmmfc2v","handle":"zhipu-glm-vision-mcp","url":"https://wellknown.network/agents/zhipu-glm-vision-mcp","links":{"self":"https://wellknown.network/agents/zhipu-glm-vision-mcp/record.json","html":"https://wellknown.network/agents/zhipu-glm-vision-mcp","markdown":"https://wellknown.network/agents/zhipu-glm-vision-mcp/record.md","api":"https://wellknown.network/api/v1/agents/zhipu-glm-vision-mcp","status":"https://wellknown.network/api/v1/agents/zhipu-glm-vision-mcp/status","claim":"https://wellknown.network/agents/zhipu-glm-vision-mcp/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/zhipu-glm-vision-mcp/claim.json","badge":"https://wellknown.network/agents/zhipu-glm-vision-mcp/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/zhipu-glm-vision-mcp/history","tools":"https://wellknown.network/api/v1/agents/zhipu-glm-vision-mcp/tools"},"ard":{"identifier":"urn:air::server:zhipu-glm-vision-mcp","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"zhipu-glm-vision-mcp","summary":"MCP server for GLM-4.6V-Flash multimodal understanding (image/video/file) via Zhipu API","description":"# zhipu-glm-vision-mcp\n\n基于智谱 GLM-4.6V-Flash（免费多模态视觉模型）的多模态理解 MCP 服务，支持图片、视频、文件三类理解能力，并支持思考模式（thinking）开关。\n\nGLM-4.6V-Flash 上下文窗口 128K，在视觉理解精度上达到同参数规模 SOTA，原生支持多模态 Function Calling。本服务通过智谱 OpenAI 兼容接口接入，遵循 **MCP v2 协议**（`mcp>=2.0.0` + `MCPServer`）。\n\n## 使用\n\n```bash\nuvx zhipu-glm-vision-mcp\n```\n\n需要设置环境变量 `ZHIPU_API_KEY`（在 [智谱开放平台](https://open.bigmodel.cn) 免费获取）。\n\n可选环境变量：\n\n| 变量 | 说明 | 默认 |\n|---|---|---|\n| `ZHIPU_API_KEY` | 智谱 API Key（必需） | - |\n| `GLM_MODEL` | 模型名称 | `glm-4.6v-flash` |\n| `GLM_THINKING` | 思考模式默认，取值与工具 `thinking` 参数一致（`disabled` / `enabled` / `enabled_with_reasoning`；兼容布尔写法 `true`/`false`） | `disabled` |\n\nCLI 参数：`--model <名称>` 覆盖模型、`--transport/-t stdio|streamable-http|sse`（默认 stdio）、`--thinking` / `--no-thinking` / `--reasoning` 设置思考默认（分别对应 `enabled` / `disabled` / `enabled_with_reasoning`）。\n\n## 工具\n\n每个工具均接受**列表参数**（传 1 个即单条分析，传多个即在同一上下文内对比/关联分析），并支持 `thinking` 思考开关与 `prompt` 自定义提问。\n\n> 本地 Base64 传入**仅图片支持**（官方契约）；视频/文件仅支持公网 URL 传入（`file_url` 官方明确不支持 Base64）。因此视频/文件只提供 `*_url` 工具。\n\n### 图片理解\n- `analyze_image(file_paths: list, prompt?, thinking?)` - 分析 1..50 个本地图片路径（JPEG/PNG，每张 Base64 ≤5MB、≤6000×6000）\n- `analyze_image_url(urls: list, prompt?, thinking?)` - 分析 1..50 个公网图片 URL（JPEG/PNG）\n\n### 视频理解\n- `analyze_video_url(urls: list, prompt?, thinking?)` - 分析 1..N 个公网视频 URL（MP4/MKV/MOV，每个 ≤200MB，数量不设上限）\n\n### 文件/文档理解\n- `analyze_file_url(urls: list, prompt?, thinking?)` - 分析 1..50 个公网文件 URL（PDF/TXT/Word/DOCX/JSONL/XLSX/PPTX），支持多文档跨文档对比\n\n### 可选参数\n\n每个工具都接受以下**可选参数**：\n\n| 参数 | 类型 | 说明 | 默认 |\n|---|---|---|---|\n| `prompt` | string | 自定义提问，覆盖默认提示词 | 按模态的默认中文提示词（描述/分析/总结） |\n| `thinking` | string | 三态思考模式，见下 | 回落 `GLM_THINKING` 环境变量，再默认 `disabled` |\n\n**必选参数**：\n- `file_paths`：本地图片路径列表，用于 `analyze_image`\n- `urls`：公网 URL 列表，用于 `analyze_image_url` / `analyze_video_url` / `analyze_file_url`\n\n**`thinking` 三态**：\n\n| 取值 | 行为 |\n|---|---|\n| `\"disabled\"` | 关闭思考，只返回最终答案 |\n| `\"enabled\"` | 开启思考，不返回思考链 |\n| `\"enabled_with_reasoning\"` | 开启思考，并返回思考链（`【思考过程】…\\n【回答】…`） |\n\n**优先级**：单次调用 `thinking` …","publisher":null,"homepage":null,"repository":null,"version":"0.1.3","license":null,"protocols":["mcp"],"tags":["glm","mcp","multimodal","vision","zhipu"],"pricing":null,"endpoints":[{"url":"pypi:zhipu-glm-vision-mcp","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","summary":"pypi","version":"pypi","description":"pypi"}},"derived":{"capabilities":[{"slug":"media.image-understanding","name":"Image Understanding","confidence":1,"provenance":"declared"}],"categories":["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":"zhipu-glm-vision-mcp","registry":"pypi","observedAt":"2026-10-23T00:27:57.455Z","publishedAt":"2026-08-02T09:04:32.921031Z","latestVersion":"0.1.3"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"zhipu-glm-vision-mcp","url":"https://pypi.org/project/zhipu-glm-vision-mcp/","firstSeenAt":"2026-09-10T16:23:54.952Z","fetchedAt":"2026-10-23T00:26:46.761Z","normalizedAt":"2026-10-23T00:26:46.761Z"}]},"firstSeenAt":"2026-09-10T16:23:54.952Z","updatedAt":"2026-10-23T00:27:57.455Z"}