{
  "markdown": "# AI Hive AIGC MCP Server\n\n<!-- mcp-name: io.github.wubin1836/ai-hive-aigc -->\n\nAI Hive AIGC is a Streamable HTTP MCP server for commercial image, video, e-commerce, marketing and advertising generation. One server exposes model discovery, media upload, image generation, video generation, e-commerce workflows, advertising/TVC workflows and task polling.\n\nAI Hive AIGC MCP 为中文用户和全球开发者提供统一的图片与视频生成入口，覆盖 Nano Banana Pro、GPT Image 2、Seedream、Seedance、MiniMax H3、Wan、HappyHorse，以及电商主图、商品详情页、Listing、PDP、广告、TVC、带货、种草、短剧和漫剧场景。\n\n## MCP endpoint\n\n```text\nhttps://ai-hive.iclip.cn/api/mcp\n```\n\nTransport: Streamable HTTP.\n\n## Authentication\n\nThe public endpoint advertises OAuth 2.1 protected-resource metadata and supports PKCE. The local reference server also supports an AI-HIVE API key in the request header:\n\n```text\nx-ai-hive-api-key: sk-api-...\n```\n\nFor local development, set `AI_HIVE_API_KEY`. The local server intentionally allows unauthenticated MCP discovery so registries can index the tool catalog; generation calls fail safely until a user provides a key.\n\n## Tools\n\n- `ai_hive_list_models`\n- `ai_hive_upload_media`\n- `ai_hive_generate_image`\n- `ai_hive_generate_ecommerce_image`\n- `ai_hive_generate_video`\n- `ai_hive_generate_ecommerce_video`\n- `ai_hive_generate_advertising_video`\n- `ai_hive_get_task`\n\nGeneration tools are non-idempotent and can incur charges. After receiving a `task_id`, query that task instead of resubmitting the same request.\n\n## Run locally\n\n```bash\nnpm install\nnpm test\nnpm start\n```\n\nHealth check: `GET /health`\n\nStatic registry card: `GET /.well-known/mcp/server-card.json`\n\nThe production MCP is hosted by AI-HIVE and is not deployed from this repository. This repository provides the public protocol reference, tool implementation, tests, Registry metadata and policy documents.\n\n## Docker\n\n```bash\ndocker build -t ai-hive-mcp .\ndocker run --rm -p 3000:3000 -e AI_HIVE_API_KEY=sk-api-example ai-hive-mcp\n```\n\n## Smithery configuration schema\n\nWhen publishing, use a session configuration field injected as an HTTP header:\n\n```json\n{\n  \"type\": \"object\",\n  \"properties\": {\n    \"apiKey\": {\n      \"type\": \"string\",\n      \"title\": \"AI Hive API Key\",\n      \"x-from\": { \"header\": \"x-ai-hive-api-key\" }\n    }\n  },\n  \"required\": [\"apiKey\"]\n}\n```\n\n## License\n\nMIT\n\n## Policies and support\n\n- [Privacy Policy](./PRIVACY.md)\n- [Terms of Service](./TERMS.md)\n- [Support](./SUPPORT.md)\n- [Security](./SECURITY.md)\n",
  "bytes": 2438,
  "sha": "2f9d6247bcb89a0df13dce414072f5998fb9144ccd5ca4e802d9415a0895b04a",
  "repo_slug": "wubin1836/ai-hive-mcp-server",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_wubin1836_ai_hive_aigc_44738eaa/readme"
}