A MCP server that provides image metadata and related tooling for AI agents. It focuses on metadata inspection without performing image processing or manipulation, exposing tools for URL-based analysis and asset discovery.
🛠️ Key Features
- image_metadata: read image metadata from a URL via HTTP headers (content-type, file size, last-modified, etag, cache info) without downloading the full image
- find_favicons: discover all favicons for a website by checking /favicon.ico, HTML link tags, and manifest.json
- extract_og_image: (excerpt mentions Open Graph, Twitter card, and Apple touch image extraction)
🚀 Use Cases
- Prepare concise image metadata for AI agents and MCP workflows
- Discover website icons and Open Graph assets for better context provisioning
- Integrate image-related metadata extraction into automated agent pipelines
⚡ Developer Benefits
- Clear, metadata-focused MCP server for agent-enabled workflows
- Tool-based endpoints with deterministic outputs
- TypeScript-based implementation aligned with MCP-server concepts
⚠️ Limitations
- No actual image processing or manipulation is performed
- Metadata extraction limited to headers and asset discovery; content not downloaded for analysis
- Scope derived from readmeExcerpt; functionality may evolve with updates