The Product
An AI helpdesk that actually reads your manuals — and now sees your photos too
What is ManualNinja?
ManualNinja is a retrieval-augmented AI assistant for technical documentation. You feed it the PDF manuals your team already relies on, and it gives every technician an instant, expert answers channel — one that never hallucinates beyond the source material, because every claim is tied to a specific page in a specific manual.
It is purpose-built for appliance installers and repair technicians who need fast, accurate, verifiable instructions on the job. Upload a manual, ask a question in plain language, and get a step-by-step answer with a direct link to the source page.
Now with AI vision: snap a photo of a device label, error screen, or nameplate, and ManualNinja will read it, identify the device, extract error codes, and find the right answer in your manuals.
Core capabilities
RAG chat assistant
AI vision (new)
Page-specific citations
Multi-tenant organizations
Google & LinkedIn login
Self-service signup
Stripe subscriptions
Tier-based limits
Bot protection
Multilingual — ask in your own language
ManualNinja automatically detects the language of your question and responds in the same language. Your manuals may be in English, but your technicians can ask in Dutch, German, French, or any other language — and get a clear, natural answer back.
Ask in any language
Manuals stay in their original language
How it works
From PDF to cited answer in four steps
1. Upload manuals
2. AI indexes
3. Ask or snap
4. Get answers
AI Vision: Snap. Identify. Solve.
Available on Business and Enterprise plans
A technician encounters an unfamiliar error code on a device display. Instead of searching through manuals, they simply snap a photo with their phone and upload it to the chat.
ManualNinja’s AI vision model analyzes the image and extracts:
- Device type — what kind of equipment is this
- Brand & model — from nameplates and labels
- Error codes — from digital displays and LED patterns
- Technical specs — voltage, wattage, pressure, etc.
- All visible text — full OCR of labels and stickers
The AI then automatically searches your manual library using this information and provides a targeted answer with the right repair procedure.
Why it matters
Field technicians often can’t identify a device model from memory, especially with older equipment or rebranded units. A photo eliminates the guesswork:
- No more calling the office to look up a model number
- Error codes are read automatically — no manual transcription errors
- Nameplate data (serial, voltage, production date) is captured instantly
- The AI matches the photo data to the right manual page
- Works with any device that has a label, screen, or nameplate
It’s like having a senior technician looking over your shoulder, who happens to have memorized every manual in your library.
Simple, transparent pricing
Start free. Upgrade when you need more. Cancel anytime.
Free
5 questions/mo
1 manual
100 MB storage
1,500 chars/question
1 user
Google/LinkedIn login
No image upload
Basic
500 questions/mo
Unlimited manuals
100 MB storage
1,500 chars/question
Unlimited users
No image upload
Pro
2,000 questions/mo
Unlimited manuals
500 MB storage
2,500 chars/question
Unlimited users
No image upload
Business
5,000 questions/mo
Unlimited manuals
2 GB storage
5,000 chars/question
Unlimited users
✓ Image upload & AI vision
Enterprise
Unlimited questions
Unlimited manuals
10 GB storage
15,000 chars/question
Unlimited users
✓ Image upload & AI vision
Ready to give your team instant answers?
Start free today. Upload your first manual and ask your first question in under 2 minutes.