AMANDA/AI
AI TOOLS

Image Intelligence

AI-powered tools built into the Amanda platform. Every image uploaded gets smarter — tagged, described, resized, and ready to serve anywhere via API.

1 tool live
·
5 in pipeline
·v0.2 → v0.4 roadmap
01
Upload
Image lands in Vercel Blob
02
Analyze
Vision model reads the image
03
Enhance
Tags, descriptions, EXIF written
04
Serve
Enriched asset available via API
AVAILABLE NOW● live
● live
Resize Engine
Resize any image to exact social media dimensions. Instagram, X, LinkedIn, TikTok, YouTube and more — all presets built in.
Sharp · POST /api/resize
Open tool →
IN THE PIPELINEEach slot is ready to receive code — flip status to 'ready' when built
v0.2
Auto Enhance
Auto-tag images by content, generate descriptions, enrich EXIF data, and assign categories — triggered automatically on every upload.
writes →tags[]descriptioncategoryexif
60%
OpenAI Vision · POST /api/enhance
v0.2
Image Describe
Generate rich alt text, captions, and SEO-ready descriptions from any image. Output feeds directly into asset metadata.
writes →altTextcaptionseoDescription
55%
OpenAI GPT-4o Vision · POST /api/describe
v0.3
AI Rerender
Upscale, style transfer, and creative rerender. Transform RAW shots into polished outputs using generative models.
writes →rerenderUrlmodelprompt
30%
Replicate · fal.ai · POST /api/rerender
v0.3
Background Removal
One-click background removal for portraits, products, and lifestyle shots. Returns PNG with full transparency preserved.
writes →bgRemovedUrlmaskUrl
25%
Replicate REMBG · POST /api/background
v0.4
Social Pack
One image in → full social media pack out. Every platform size generated in a single job, returned as a zip.
writes →socialPack{}packUrlformats[]
10%
Sharp batch · POST /api/social-pack
🔧
Adding a new AI tool
Add an entry to TOOLS[] in app/ai/page.tsx → build the API route at app/api/[tool]/route.ts → flip status: 'ready' and set href. The card goes live automatically.