Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when auditing Open Graph tags or generating og:image assets. Applies to any page that may be shared on social networks or messaging platforms.
.claude/skills/thedaviddias-og-image-size/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-22 | ✗→✓ | ▲ Improved | -37% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | -10% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | -6% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | -4% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | -18% | 0% |
An incorrectly sized og:image is displayed poorly or not at all on social platforms—Facebook, Twitter/X, LinkedIn each crop or skip images that don't meet their requirements, resulting in blank or distorted link previews.
Find the og:image meta tag in <head> and fetch the image URL. Check the image dimensions and file size. Flag if width < 1200px, height < 630px, aspect ratio differs significantly from 1.91:1, or file size exceeds 5 MB.
Generate or resize the og:image to exactly 1200×630 pixels. Export as JPEG at 80–85% quality to balance visual quality and file size. Reference the absolute HTTPS URL in the og:image meta tag. Add og:image:width and og:image:height tags to help platforms render the image without downloading it first.
Social platforms use og:image to construct the visual card shown when a URL is shared. An image too small may be skipped or shown as a thumbnail. An image with the wrong aspect ratio gets cropped, often cutting off important content. The recommended 1200×630 size ensures optimal display across Facebook, LinkedIn, Slack, and iMessage.
Check the og:image meta tag value. Fetch the image and verify: (1) dimensions are at least 1200×630 px, (2) aspect ratio is approximately 1.91:1, (3) file size is under 5 MB, (4) URL is absolute HTTPS, (5) og:image:width and og:image:height tags match actual dimensions. Flag relative URLs or images smaller than 600×315 px.
For full implementation details, code examples, and framework-specific guidance, see references/rule.md.
Rule page: https://frontendchecklist.io/en/rules/seo/og-image-size
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-04 | pass→pass | 8,734 | 5,751 | -34% | 1 | 1 | 0% | 1,763 | 1,592 | -10% | 0 | 0 | — |
case-01 | pass→pass | 9,566 | 5,710 | -40% | 1 | 1 | 0% | 1,778 | 1,669 | -6% | 0 | 0 | — |
case-02 | pass→pass | 7,250 | 4,469 | -38% | 1 | 1 | 0% | 1,420 | 1,362 | -4% | 0 | 0 | — |
case-03 | pass→pass | 9,240 | 5,038 | -45% | 1 | 1 | 0% | 1,835 | 1,502 | -18% | 0 | 0 | — |
case-05 | pass→pass | 10,745 | 6,180 | -42% | 1 | 1 | 0% | 1,816 | 1,570 | -14% | 0 | 0 | — |
case-06 | pass→pass | 10,405 | 4,851 | -53% | 1 | 1 | 0% | 1,773 | 1,421 | -20% | 0 | 0 | — |
case-07 | pass→pass | 10,822 | 6,078 | -44% | 1 | 1 | 0% | 1,777 | 1,650 | -7% | 0 | 0 | — |
case-08 | pass→pass | 9,744 | 2,670 | -73% | 1 | 1 | 0% | 1,547 | 844 | -45% | 0 | 0 | — |
case-09 | pass→pass | 9,209 | 3,755 | -59% | 1 | 1 | 0% | 1,695 | 1,288 | -24% | 0 | 0 | — |
case-10 | pass→pass | 9,801 | 5,502 | -44% | 1 | 1 | 0% | 1,918 | 1,741 | -9% | 0 | 0 | — |
case-11 | pass→pass | 5,158 | 3,954 | -23% | 1 | 1 | 0% | 1,113 | 1,359 | +22% | 0 | 0 | — |
case-12 | pass→pass | 6,314 | 4,229 | -33% | 1 | 1 | 0% | 1,222 | 1,428 | +17% | 0 | 0 | — |
case-17 | pass→pass | 11,205 | 8,547 | -24% | 1 | 1 | 0% | 2,193 | 2,301 | +5% | 0 | 0 | — |
case-13 | pass→pass | 9,089 | 4,135 | -55% | 1 | 1 | 0% | 1,686 | 1,365 | -19% | 0 | 0 | — |
case-14 | pass→pass | 8,176 | 5,231 | -36% | 1 | 1 | 0% | 1,490 | 1,560 | +5% | 0 | 0 | — |
case-15 | pass→pass | 8,686 | 4,834 | -44% | 1 | 1 | 0% | 1,686 | 1,495 | -11% | 0 | 0 | — |
case-16 | pass→pass | 10,783 | 6,456 | -40% | 1 | 1 | 0% | 1,965 | 1,653 | -16% | 0 | 0 | — |
case-18 | pass→pass | 11,245 | 3,234 | -71% | 1 | 1 | 0% | 1,813 | 1,139 | -37% | 0 | 0 | — |
case-19 | pass→pass | 4,319 | 4,285 | -1% | 1 | 1 | 0% | 892 | 1,352 | +52% | 0 | 0 | — |
case-20 | pass→pass | 7,730 | 5,528 | -28% | 1 | 1 | 0% | 1,493 | 1,560 | +4% | 0 | 0 | — |
case-21 | pass→pass | 12,040 | 7,510 | -38% | 1 | 1 | 0% | 2,143 | 1,904 | -11% | 0 | 0 | — |
case-22 | fail→pass | 11,070 | 4,194 | -62% | 1 | 1 | 0% | 2,252 | 1,426 | -37% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +5 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.