seo-images analyzes the images on a page for SEO and speed. It checks alt text, tiered file size thresholds, the format a browser receives, srcset and sizes, lazy loading, fetchpriority for the LCP image and width and height to prevent layout shift.
It also optimizes image files: WebP and AVIF conversion, responsive variants, IPTC and XMP metadata, and the DigitalSourceType label that Merchant Center expects on AI-generated product images. With DataForSEO available it compares your images with the image results in a SERP.
What You Get
- An audit table counting missing alt text, oversized images, wrong formats and missing dimensions
- A prioritized optimization list with estimated savings per image
- WebP or AVIF files, responsive variants and picture element markup
- IPTC, XMP and AI source labels written into image files
Common Use Cases
- Find the images that slow a landing page and hurt LCP
- Bulk convert a folder of JPEGs to WebP with metadata kept
- Label AI-generated product photos before a Merchant Center upload
- Compare your image presence with the image SERP for a keyword
What Makes It Unique
It judges the format a server really delivers by probing each image with a browser Accept header, so a .jpg URL that serves WebP is not flagged. It also separates what matters for Google Images (alt text, file name, page context) from what does not (EXIF, IPTC keywords), and covers AI-image labeling.
Under The Hood
The skill is written instructions. It runs the plugin’s Python helper through a launcher to parse the page HTML and, with a probe option, to fetch each image over the network. For file optimization it tells the agent to run exiftool, cwebp, ImageMagick and ffmpeg on the machine, which rewrites image files and metadata, and an IPTC helper script can audit or inject labels. An optional DataForSEO call fetches SERP data. The frontmatter declares none of this access. Fetched pages and API responses are to be treated as untrusted data.