An evidence-based skill for Codex and Claude Code that validates crawler-visible Open Graph and Twitter Card metadata, repairs framework-native metadata, and creates branded 1200×630 preview assets when needed.
fix-social-link-previews is a AI-agent skill created to make a real workflow clearer, faster, and easier to trust. It turns a focused product idea into something people can use, evaluate, and understand without needing to inspect the source code.
Why fix-social-link-previews exists
A page can look correct in a browser while link-card crawlers see missing/client-only metadata, an inaccessible/small image, or stale platform cache.
The goal is not simply to add another tool. fix-social-link-previews is designed to reduce the friction around that problem and give people a more dependable way to complete the work.
What it delivers
- Agents can diagnose missing versus cached cards.
- Repair incomplete/relative/client-only metadata.
- Create a branded large-image asset when needed.
- Verify local implementation and deployed crawler eligibility without promising a platform-specific layout.
The experience
Audits, creates, repairs, and verifies large-image social previews for web apps and websites.
The project brings the important parts of the workflow into one coherent experience. Its interface and behavior are shaped around the people using it, while the implementation stays focused on predictable results and honest feedback.
What to know
Every product has boundaries. These are the most important considerations for fix-social-link-previews:
- Proprietary social platforms' selection and cache algorithms cannot be simulated or controlled.
- Metadata must be crawler-visible and image URLs remotely reachable.
- Only PNG, JPEG, and supported WebP encodings yield deterministic dimension checks.
These limitations are presented openly so visitors can understand where the project works well, what it depends on, and what may change in future versions.
Current status
Published npm package v0.3.0. The current release demonstrates the core value of the project and the main end-to-end experience.
My role was agent workflow design and automation engineering, covering agent workflow, validation automation, and packaging. The related case study explains the decisions and delivery process in greater depth for visitors who want the engineering story behind the product.