Changelog
What's new in folii.ai.
1.3.0
April 16, 2026
Blog & writing
Features
- Blog posts on your portfolio — write essays, project deep-dives, and announcements that live at folii.ai/your-username/blog. Each post is published independently and gets a clean reader page with title, date, read time, and tags.
- Chat-driven writing — ask the AI to draft a post, expand a section, change tone, or add tags. The blog chat keeps per-post history so you can iterate over multiple turns. It also reads your portfolio for context, so "write a post about my Stripe integration project" actually pulls from your project list.
- Drafts vs published — every post starts as a draft. Edit as long as you want; only "Published" posts show up at /your-username/blog. Toggle anytime.
- Tags and excerpts — tag posts with topic keywords (shown as pills on the listing) and write a short excerpt that previews the post on the index.
- Syntax-highlighted code blocks — fenced code in your markdown gets GitHub-style highlighting (dark and light themes auto-switch with the reader's theme).
- Autosave with unsaved-changes guard — typing autosaves in the background; navigating away with unsaved edits warns you first.
Improvements
- Confirmation dialogs for destructive actions in the editor (delete project, reset portfolio, etc.) so you don't lose work to a stray click.
- Logo and favicon refresh — the folii mark now appears in the browser tab.
- Chat route timeout extended to 120s for longer edits on large portfolios.
Fixes
- GitHub OAuth redirect now derives from the request origin, so importing repos works on production deploys, not just localhost.
- GitHub import button no longer freezes the browser when triggering the confirm dialog.
- Public blog post pages no longer crash with a "runSync finished async" error when rendering syntax-highlighted code blocks.
- Blog chat correctly shows the assistant reply when the API succeeds (was previously surfacing "No reply returned" on success).
- Blog editor reflects chat-driven post updates when you haven't typed anything, instead of appearing stale.
1.2.0
April 12, 2026
GitHub OAuth, profile import & unified Import menu
Features
- GitHub OAuth integration — connect your GitHub account to browse and import repos directly. Select up to 5 repositories and folii creates polished project entries from each repo's metadata and README.
- GitHub profile import — one click to fill your About Me and Skills. folii scans your repos, aggregates languages and topics, and generates a bio, tagline, and skill categories from your actual code.
- Unified Import menu — Resume, GitHub, and LinkedIn are now organized under a single Import button with a clean dropdown. Resume upload moved from the top bar into the menu.
- Importing docs page — new documentation at /docs/importing covering all three import methods with step-by-step instructions.
Improvements
- Import button uses primary blue styling so it stands out as the first action for new users.
- LLM output coercion handles null values, bare URLs, and array-to-string mismatches across all import flows.
Fixes
- Twitter renamed to X on the contact page.
- Theme selection dropdown now renders above the light/dark mode toggle.
1.1.0
April 12, 2026
LinkedIn import
Features
- LinkedIn profile import — export your LinkedIn profile as PDF and upload it to folii. AI parses your experience, education, skills, and bio into a polished portfolio. Step-by-step instructions included in the UI.
1.0.0
April 12, 2026
v1 launch
Features
- GitHub project import — paste a public repo URL in chat or use the Import button in the topbar. folii pulls the name, description, language, stars, and topics from the GitHub API and writes a polished project entry.
- Improved landing page — new features section, "Why folii" differentiator, sharper hero copy, and stronger final CTA.
- Documentation site at /docs — getting started guide, AI editing tips, themes guide, publishing walkthrough, and FAQ.
- Changelog page at /changelog.
- Sign out button and dark/light theme toggle on the landing page.
- Sentry error tracking — production error monitoring with automatic Linear ticket creation.
Improvements
- Rotating chat placeholder hints — cycles through example prompts every 4 seconds to help new users.
- Better rate limit error messages — shows remaining count instead of misleading "try again in 60s".
- API timeouts — 45s on OpenAI calls, 5s on GitHub fetch, preventing silent function hangs.
- Error logging — LLM failures now log to console for Vercel log visibility.
- Smarter error messages — distinguishes "AI returned bad format" from "AI unavailable".
Fixes
- JSON editor saves no longer create "Applied direct JSON edit" chat messages.
- Removed split/focus layout toggle — editor always uses split view for a cleaner experience.
0.0.1
April 12, 2026
Initial release
Features
- Resume upload and AI parsing — drop a PDF and get a structured portfolio in seconds.
- AI chat editor — edit your portfolio in natural language. Rewrite your bio, add projects, remove sections, all through conversation.
- Live preview — see every change in real time as you edit.
- One-click publish — your portfolio goes live at folii.ai/username instantly.
- Publish and unpublish toggle — take your site offline and back with one click.
- 43 theme presets — dark and light variants for every theme, sourced from the tweakcn community.
- Dark/light mode toggle — visitors can switch between dark and light on your published site.
- Smart chat suggestions — context-aware chips that suggest edits based on what's missing in your portfolio.
- Undo/revert — roll back any AI edit from the chat history.
- JSON editor — power users can edit the raw portfolio data directly.
- Section hiding — hide Experience, Skills, Projects, or Contact from the nav and published page.
- Project images — upload screenshots for your project cards.
- Profile avatar — upload and crop a profile photo.
- SEO meta tags — Open Graph and Twitter card tags with a dynamically generated preview image.
- Mobile responsive — editor and published portfolios work on all screen sizes.
Improvements
- Rate limiting on AI chat and resume parsing to keep things fair.
- Inline error messages with retry button when AI calls fail.
- Navigation between editor and landing page.